Skip to main content

Full text of "A study of prolonged fasting"

See other formats


w 



/< 




A STUDY OP PROLONGED FASTING 



BY 



FRANCIS GANO BENEDICT 




WASHINGTON, D. C. 

Published by the Carnegie Institution op Washington 

1915 






Carnegie Institution of Washington 
Publication No. 203 



PRESS OF GIBSON BROTHERS, INC. 
WASHINGTON, D. C. 



PREFACE. 

The research reported in this book on the metabolism during pro- 
longed fasting is a continuation and amplification of the investiga- 
tions reported in "The influence of inanition on metabolism" (Carnegie 
Institution of Washington Publication No. 77, 1907). 

The opportunity to conduct this series of scientific observations on a 
man living for 31 days without food and drinking only distilled water 
would have been of little value without the cooperation of a large 
number of scientific associates and computers. Certain co-workers 
kindly assumed the responsibility not only for the accumulation of the 
data but also for the preparation of a report of their respective findings. 
In this book special reports are made by Dr. H. W. Goodall on the 
physical condition of the subject during the fast, his subjective impres- 
sions and mental attitude toward the fast, and the microscopy of the 
urine and the tests for albumin; by Dr. J. E. Ash on the blood; by Dr. 
H. S. Langfeld on the psycho-physiology of the fast; by Dr. A. I. Kendall 
on bacterial intestinal flora; and by Mr. H. L. Higgins on alveolar air. 

Aside from those who shared directly in the responsibility of the 
studies, I am indebted to numerous scientific authorities for counsel and 
advice, both during the experiment and during the preparation of the 
material for publication. Those not specifically mentioned in the text 
are Professors Luciani of Rome, Fano of Florence, Zuntz of Berlin, 
Tangl of Budapest, Tigerstedt of Helsingf ors, and Lusk of New York. 

In no undertaking of the Nutrition Laboratory have the concentration 
and the unification of resources and assistants been so intensely applied 
and to the whole staff of the Laboratory my warmest thanks are due. 
Their interest and conscientious, painstaking work alone made sure 
the collection of the data reported in the following pages. The labor 
of the final preparation of the material has fallen in no small part into 
the excellent hands of my editorial associates, Mr. W. H. Leslie and 
Miss A. N. Darling. 

Nutrition Laboratory of the Carnegie Institution of Washington, 

Boston, Mass., July 29, 191 4. 
3 



CONTENTS. 



PAGE. 

Introduction 11 

Previous observations of prolonged fasts 13 

Observations on Succi 16 

Research on metabolism in prolonged fasting at the Nutrition Laboratory 19 

Problems to be studied 19 

Selection of subject 20 

Proofs of physical fitness 21 

Autobiographical notes 22 

General characteristics of subject 28 

General history of fasting experiment 29 

Program for research 31 

Daily records of fasting experiment 32 

Preliminary period 32 

Fasting period 37 

Re-alimentation period 49 

Physical condition of the subject during the fast 53 

Results of physical examination 54 

Summary as to physical condition 63 

Photographic study of subject 66 

Anthropometric measurements 67 

Body-weight 69 

Routine of observations 69 

Daily losses in body-weight in fasting experiments 71 

Total loss in body-weight *. 80 

Analysis of losses in body-weight 82 

Insensible perspiration 83 

Drinking water 85 

Body-temperature 88 

Changes in temperature rhythm 89 

Observations of the body-temperature in the night period 92 

Average body-temperature 92 

Range in body-temperature 94 

Observations of the body-temperature in the day period 95 

Constancy in body-temperature at a given hour 97 

Pulse-rate 99 

Records of pulse-rate obtained in earlier fasting experiments 100 

Records of pulse-rate obtained in the experiment with subject L 103 

Pulse-rate in the night periods 106 

Pulse-rate in the day periods 110 

Comparison of pulse records obtained in experiments with the bed calori- 
meter and the respiration apparatus 112 

Influence of body position 115 

Influence of the work of writing 116 

Influence of breathing an oxygen-rich atmosphere 116 

Diurnal rhythm 117 

Irritability of the heart 117 

Blood pressure ' 119 

The blood 124 

Correlation of literature 124 

Erythrocytes 125 

Haemoglobin 132 

Leucocytes 136 

Physico-chemical changes 144 

Observations on L's blood 148 

Discussion and conclusions 156 

Mechanics of respiration 158 

Typical graphic records of respiration 158 

Method of calculating the total ventilation of the lungs 160 

Method of calculating the volume per inspiration 161 

Results of observations on the mechanics of respiration 162 

Respiration-rate 163 

Ventilation of the lungs per minute 164 

5 



6 A STUDY OF PROLONGED FASTING. 

PAGE. 

Research on metabolism in prolonged fasting at the Nutrition Laboratory — Continued. 
Mechanics of respiration — Continued. 

Results of observations on the mechanics of respiration — Continued. 

Volume per inspiration 164 

Influence of changes in body position 165 

Influence of the work of writing 165 

Influence of breathing an oxygen-rich atmosphere 166 

Maximum expiration of the lungs 166 

Alveolar air 168 

Significance of alveolar air 168 

Methods of determining the alveolar air 169 

Haldane method 169 

Plesch method 171 

Method of calculating alveolar air from respiration experiments 172 

Conditions of taking alveolar-air samples " 174 

Discussion of results 175 

Size of dead space in fasting 175 

Difference in mechanics of respiration in morning and evening 178 

Significance of change in the alveolar air during the fast 180 

Conclusions 181 

Subjective impressions and mental attitude toward the fast 182 

Subjective impressions 182 

Mental attitude of the subject toward the fast 187 

The psycho-physiology of a fast 191 

Memory for words 193 

Tapping tests 194 

Strength tests 196 

Tactual-space threshold 198 

Rote memory for digits 199 

Association tests 200 

Cancellation test 206 

Visual acuity 207 

Later tests 208 

Correlations 211 

General summary and conclusions 212 

Appendix I. Dreams 222 

Appendix II. Complete series of association tests 222 

Feces 230 

Observations upon the bacterial intestinal flora of a starving man 232 

Excretion through the skin 233 

Urine 236 

General routine of collection and sampling 236 

Composition of the urine prior to the fasting experiment 238 

Physical characteristics of the fasting urine 238 

Volume of urine 240 

Specific gravity 242 

Total solids 243 

Day and night urines 245 

Chemical constituents of fasting urine 247 

Total nitrogen 247 

Comparison of total nitrogen excretion of L. with that of other fasting 

subjects 247 

Daily excretion of nitrogen 250 

Nitrogen excretion per kilogram of body-weight 252 

Comparison of methods for determining total nitrogen and ammonia- 
nitrogen 253 

The partition of the nitrogen excretion 254 

Urea 254 

Ammonia 257 

Uric acid 259 

Creatinine 262 

Rest-nitrogen 268 

Acid radicles 268 

Chlorine 268 



CONTENTS. / 

Research on metabolism in prolonged fasting at the Nutrition Laboratory — Continued. page. 
Urine — Continued. 

Chemical constituents of fasting urine — Continued. 

Phosphorus 273 

Sulphur 277 

Total acidity 281 

/8-oxybutyric acid 282 

Mineral metabolism 285 

Relationships of the mineral constituents 287 

Reducing power 291 

Carbon in urine 293 

Carbon-nitrogen ratio 295 

Energy of urine 297 

Calorie-nitrogen ratio 297 

Calorie-carbon ratio 298 

Microscopy of urine and tests for albumin 300 

Detailed results 300 

Summary 302 

The respiratory exchange 304 

Apparatus and methods used in the calorimeter experiments 305 

Absorption of water-vapor and carbon dioxide 305 

Analysis of chamber air at the end of periods 306 

Tension equalizer 309 

Argon in oxygen from liquid air 310 

Graphic registration of degree of muscular repose of subject inside the respi- 
ration calorimeter 311 

Methods used in experiments with the respiration apparatus 315 

Studies with the bed calorimeter 320 

Atmospheric conditions inside the chamber 320 

Measurement of the respiratory exchange inside the bed calorimeter 322 

Periodic changes in the metabolism 323 

Total metabolism 327 

Respiratory quotient 330 

Relationships of pulse-rate, body-temperature, and metabolism 331 

Studies with the universal respiration apparatus 333 

Variations in the metabolism as the fast progressed 335 

Relationship between the pulse-rate and the metabolism 336 

Diurnal variations in metabolism 337 

External influences upon metabolism 338 

Effect of changes in body position 338 

Influence of the work of writing 340 

Influence of breathing an oxygen-rich atmosphere 341 

Influence of sleep 343 

Metabolism per unit of weight and surface 351 

Metabolism per kilogram of body-weight 353 

Metabolism per kilogram of body-weight in calorimeter experiments . . 359 
Metabolism per kilogram of body-weight in respiration-apparatus ex- 
periments 362 

Conclusions regarding the metabolism per kilogram of body-weight . . . 364 

Metabolism per square meter of body-surface 366 

Metabolism per square meter of body-surface in the calorimeter experi- 
ments 369 

Metabolism per square meter of body-surface in the respiration-appa- 

tus experiments 370 

Conclusions regarding the metabolism per square meter of body-surface . 370 
Summary of results regarding the metabolism per kilogram of body-weight 

and per square meter of body-surface 372 

Elimination of water through lungs and skin 373 

Calorimetry 379 

Direct calorimetry 379 

Indirect calorimetry 384 

Balance of income and outgo 392 

Total katabolism per 24 hours 392 

Daily activity 393 

Total carbon-dioxide production and oxygen consumption per 24 hours .... 395 



8 A STUDY OF PROLONGED FASTING. 

PAOE. 

Research on metabolism in prolonged fasting at the Nutrition Laboratory — Continued. 
Balance of income and outgo — Continued. 

Character of the katabolism 399 

Protein katabolism 400 

Apportionment of non-protein katabolism between carbohydrate and fat. . . 401 
Carbon dioxide produced and oxygen consumed in the katabolism of 

carbohydrate and fat 402 

Significance of the non-protein respiratory quotients 403 

Energy derived from katabolism of carbohydrate and fat 405 

Amounts of carbohydrate and of fat katabolized 406 

Loss of water from the body 407 

Loss of preformed water 408 

Total loss of original body-substance 412 

Total energy loss 413 



ILLUSTRATIONS 



PAGE. 

Plate 1. A. Characteristic pose of L., sitting in the balcony, during the day, writing at 
his desk. B. Use of universal respiration apparatus for studying the respira- 
tory exchange while writing 10 

Plate 2. C. Respiration experiment made by T. M. Carpenter on the universal respiration 

apparatus. D. Weighing the subject on the thirty-first day of the fast. . . 18 

Plate 3. E. The subject L. ascending the stairs of the balcony on the thirty-first day of 

the fast. F. Clinical examination by Dr. H. W. Goodall 30 

Plate 4. Views of subject Levanzin on first day of 31-day fast 64 

Plate 5. Views of subject Levanzin on last day of 31-day fast 64 

Fig. 1. Body-weight curves for fasting experiments with Succi 74 

2. Body-weight curve for Levanzin 75 

3. Body-weight curves for prolonged fasting experiments with dogs 77 

4. Body-temperature curves during the night and early morning for the second and 

fourth to eighth days of fast 90 

5. Body-temperature curves during the night and early morning for the ninth to the 

sixteenth days of fast 91 

6. Body-temperature curves during the night and early morning for the seventeenth 

to twenty-second days of fast 92 

7. Body-temperature curves during the night and early morning for twenty-third to 

twenty-ninth days of fast 93 

8. Body-temperature curves during the night and early morning for thirtieth and 

thirty-first days of fast and second and third days with food 94 

9. Body-temperature curves for approximately 24 hours on twenty-fourth and 

twenty-fifth days of fast 96 

10. Body-temperature curves showing change from lying to sitting position 97 

11. Body-temperature curve showing change from lying to sitting position 97 

12. Pulse-rate chart of subject L for days preceding fast 104 

13. Pulse-rate chart of subject L. for first to fifth days of fast 105 

14. Pulse-rate chart of subject L. for sixth to eleventh days of fast 106 

15. Pulse-rate chart of subject L. for twelfth to eighteenth days of fast 107 

16. Pulse-rate chart of subject L. for nineteenth to twenty-fifth days of fast 108 

17. Pulse-rate chart of subject L. for twenty-sixth to thirtieth days of fast 109 

18. Pulse-rate chart of subject L. for thirty-first day of fast and three subsequent 

days with food 110 

19. Chart showing blood pressure, pulse pressure, and pulse-rate of subject L 121 

20. Chart I. Relation of haemoglobin to erythrocytes. Chart II. Composite curve of 

the polynuclears compared with one of mononuclears 152 

21. Charts III and IV. Relation of total to differential leucocyte counts 153 

22. Specimen respiration curves for subject L. when lying on couch in experiments 

with the respiration apparatus 159 

23. Memory tests 193 

24. Tapping tests 195 

25. Strength tests 196 

26. Strength tests 197 

27. Strength tests 198 

28. Tactual-space threshold and visual acuity 199 

29. Free association tests 201 

30. Association tests. Reactions to verbs and nouns 201 

31. Association tests. Reactions to adjectives 202 

32. Association tests. Reactions to abstract nouns 202 

33. Reproduction tests and mean variations 203 

34. Controlled association tests 205 

35. Cancellation tests 206 

36. Specimen records of change in volume of the spirometer on the bed calorimeter 

during last 5 minutes of periods in experiment with L 310 

37. Method for obtaining graphic record of activity in bed calorimeter 312 

38. Specimen pneumograph records of movements of bed calorimeter lever mattress 

support in night experiments with L 314 

39. Schematic outline of universal respiration apparatus 316 

9 



10 A STUDY OF PROLONGED FASTING. 

PAGE. 

Fig. 40. Spirometer for studying the mechanics of ventilation 318 

41. Curves showing oxygen consumption, carbon-dioxide production, and respiratory 

quotient during night periods in the bed calorimeter for the four days preceding 

the fast and the first to the fourth days of the fast 323 

42. Curves showing oxygen consumption, carbon-dioxide production, and respiratory 

quotient during night periods in the bed calorimeter for the fifth to the fifteenth 

days of the fast 324 

43. Curves showing oxygen consumption, carbon-dioxide production, and respiratory 

quotient during night periods in the bed calorimeter for the sixteenth to the 
twenty-fourth days of the fast 325 

44. Curves showing oxygen consumption, carbon-dioxide production, and respiratory 

quotient during night periods in the bed calorimeter for the twenty-fifth to the 
thirty-first days of the fast and the second and third food days 326 

45. Complete metabolism chart of fasting dog (Awrorow No. 2) 356 

46. Complete metabolism chart of fasting dog (Awrorow No. 3) 357 

47. Metabolism chart of the most important factors measured on subject L. through- 

out the fast 416 



FASTING 



PLATE 1 




A. Characteristic pose of L. sittin e in the Balcony, during the day, writing at his deslc. 




B. Use of Universal Respiration Apparatus for studying the Respiratory Exchange while writing. 



INTRODUCTION. 

Prolonged fasting has formed a part of religious ceremony for centu- 
ries. In early times the ascetic, in his efforts to subdue all carnal 
desires, believed it necessary to withdraw from the distractions of daily 
life and to abstain either wholly or in part from food, particularly the 
flesh of animals; by thus refraining from material things, he hoped to 
be free for spiritual thought and philosophical introspection. 

Periodic fasting still constitutes a part of the rites of some religious 
bodies, particularly among the Hebrews, but in modern times a pro- 
longed fast is usually undertaken either in the hope of curing or alle- 
viating some ailment or for pecuniary gain. When a fast is resorted 
to for its supposed therapeutic value, information as to its history and 
results usually appears in one of the numerous books published by the 
advocates of peculiar dietetic regimes. When a fast is made by a 
so-called " professional faster" for pecuniary gain, the subject is exhib- 
ited to the public as an attraction to the lovers of sensational amuse- 
ments. Three decades ago such exhibitions were not uncommon and 
in many instances the subjects consented (possibly in the hope of 
increasing the interest in their performance) to more or less strictly 
controlled observations of their fasts. Not infrequently the observa- 
tions made in these professional fasting exhibitions have contributed 
materially to the sum of human knowledge, since there is an intense 
physiological interest in the vital processes during such prolonged 
abstinence from food. 

When one considers the complex activities which make up the life of 
man, it will be seen that no mechanism thus far invented approximates 
the high organization of the vital processes which are necessary to the 
life of even the simplest of the warm-blooded animals ; and yet sufficient 
experimental evidence has been accumulated to show that under normal 
conditions of life, and with similar routine, there are no marked vari- 
ations in the life processes of normal individuals. Under varying 
conditions of life, however, we find that the vital activities are carried 
on with a greater or less intensity, this being true even of the normal 
individual. We thus see that there may be definite, well-established 
planes of vital activity. For example, when the average healthy indi- 
vidual is lying in bed asleep, there is no intellectual activity and no exter- 
nal muscular activity, the vital activity being only sufficient for simple 
maintenance. When he is lying quietly in bed awake, the plane of 
vital activity is higher, and as we study the metabolism under the vary- 
ing conditions of sitting, standing, walking, and doing muscular work, 
we find an increasing intensity in the vital processes, with an increase 
in productive capacity and often an increase in efficiency. 

11 



12 A STUDY OF PROLONGED FASTING. 

The average normal man represents the mean between the two 
extremes of the emaciated, half-starved individual, disinclined to phys- 
ical or mental work, and the over-fed, obese epicure, both extremes 
being relatively low in vital activity and in productivity. Further- 
more, if we consider the metabolism under pathological conditions, we 
find even greater variations in the different levels of vital activity. 
Thus a sick person, much emaciated, lying in bed without food, and 
with subnormal temperature, has unquestionably a low cellular activity. 
On the other hand, a sick person with a high fever, even when asleep 
and without extraneous muscular activity, may have a greatly increased 
cellular activity. It will be seen, therefore, that from the standpoint 
of both normal physiology and pathology, a study of human individuals 
under different conditions and with different planes of activity is of 
fundamental importance. 

For such study it is essential to determine the basal or fundamental 
metabolism, when the activities are on a low plane, to be used as a basis 
of comparison with other values. We may ask, then, "What is the 
lowest plane of vital activity which is compatible with life?" Unques- 
tionably there have been severe pathological cases, with emaciation and 
muscular atrophy, in which life has been maintained at a plane far 
below that which can be reached by the average normal man, but it has 
been the prime object of most investigators in metabolism to concen- 
trate their efforts upon securing, with normal individuals, physiological 
values which may withstand criticism, since these constitute the only 
true basis of comparison. 

Taking into consideration the influence upon metabolism of muscular 
activity, of the ingestion of food, and the state of being awake, we may 
assert that the lowest metabolic plane would be found for an individual 
during deep sleep in bed, with complete muscular repose, and without 
food in the alimentary tract. As a matter of fact, with most people 
such a condition is usually closely approximated each day about 4 a. m. 
While in general no food is taken by an individual for about 10 or 12 
hours during the night, yet for a considerable period of time after the 
evening meal nutrients are being absorbed from the ingested food mate- 
rials and carried to different parts of the body, there to be oxidized or 
deposited. It is furthermore true that certain molecular fragments, 
probably acid in nature, maybe absorbed from the food materials which, 
when carried to the various parts of the body, may actually stimulate 
metabolism to a greater intensity, these being the so-called katabolic 
stimuli. Usually the influence of the ingestion of food ceases from 
6 to 8 or 10 hours after the meal, particularly if the food ingested is not 
protein-rich. Accordingly, for one or two hours prior to rising in the 
morning the normal man is probably living at his lowest metabolic 
plane. 



PREVIOUS OBSERVATIONS OF PROLONGED FASTS. 13 

As is well known, the normal body is liberally provided with reserve 
material, a fact which has been strikingly brought out and emphasized 
by Meltzer. 1 Consequently there is always a plethora of available 
material stored in the body for drafts in emergencies. In the normal 
life of man, the demands for nutrition are usually met by periodic 
feeding. When the demands are not met, body reserves must be drawn 
upon. Under such conditions it is of particular interest to note what 
kind of body-material is first used, the rapidity of its depletion, and the 
proportions of the various body constituents disintegrated as the drafts 
continue. It is to study these problems that observations are made 
upon fasting individuals. Furthermore, since many prominent clin- 
icians are inclined to consider disease as closely allied to the various 
stages of inanition, data secured in a study of metabolism during fasting 
have a great pathological importance for interpreting the transforma- 
tions of matter in disease. 

PREVIOUS OBSERVATIONS OF PROLONGED FASTS. 

The literature giving the results of observations during fasts has been 
reviewed at some length in a previous publication, 2 special emphasis 
being laid upon the results obtained in the earlier stages of a fast. In 
this publication it seems desirable to give a review of the longer fasts 
which have been more or less scientifically controlled and whose results 
can be considered as worthy of careful consideration. 

The longer fasts have almost without exception been made by pro- 
fessional fasters who, for purposes of exhibition, have purposed going 
without food for a definite length of time. While such a purpose would 
of itself seem to show an abnormal mental condition, yet the majority 
of professional fasters who have been used in these experiments are for 
the most part physically strong, and the results may usually be looked 
upon as of physiological importance, not complicated by pathological 
lesions of any measurable magnitude. This is particularly fortunate, 
as many fasts reported in the daily press are undertaken as a therapeutic 
measure to overcome some more or less definitely localized organic 
or functional trouble. It is obvious, however, that such experiments 
are of physiological importance when the subjects are normal individ- 
uals, voluntarily fasting under strict scientific control. 

Many professional fasters have made experiments of longer or shorter 
duration and have been studied by various investigators, but none have 
been so carefully studied and had so many experiments made with 
them and of such long duration as the Italian, Succi. Indeed, the 
classical work of Luciani on Succi emphasized perhaps more than any 

Seltzer, The Factors of Safety in Animal Structure and Animal Economy. Harvey Society 
Lectures, New York, N. Y., 1906-1907, p. 139. 

2 Benedict, Carnegie Inst. Wash. Pub. No. 77, 1907. 



14 A STUDY OF PROLONGED FASTING. 

other piece of research the importance of studying prolonged fasting. 
In this review of the literature on long fasts, therefore, brief descriptions 
of the fasts made by subjects other than Succi will first be given chrono- 
logically, these being followed by descriptions of experiments made 
with the Italian subject. Such discussion of the results as may be 
necessary will be reserved for later chapters. 

Observations by Paton and Stockman. 1 — An experiment was made in 
the fall of 1888 by Paton and Stockman on the professional faster 
Jacques and continued for 30 days. The body-weight was recorded, 
but unfortunately the urine was analyzed by the old hypobromite 
method. Furthermore, the values for total nitrogen output were un- 
doubtedly disturbed by the singular fact that the subject drank from 
60 to 300 c.c. of his own urine each day. Since the volume of fluid 
taken per day varied greatly, the body-weight fluctuated considerably, 
actual gains in weight being shown on some days. No feces were 
passed during the fast. 

Observations by Lehmann, Mueller, Munk, Senator, and Zuntz} — 
Although this research was hardly long enough to be called a study of 
prolonged fasting, the two experiments made by Lehmann, Mueller, 
Munk, Senator, and Zuntz, one on Cetti of 10 days and one on Breit- 
haupt of 6 days, present a study of metabolism during fasting which 
has never been excelled in accuracy for this length of time. The 
experimental plan adopted in this research has been followed with but 
minor changes by practically all succeeding investigators. It was the 
intention to continue the experiments with these subjects for 20 
or 30 days, but they were unavoidably shortened, owing to the condi- 
tion of the subjects. The experiment on Cetti was made in March 
1887, and the observations secured in this experiment were of such 
importance that the experimenters took advantage of an opportunity 
occurring in March 1888 to make an experiment with the professional 
faster Breithaupt. Unfortunately this experiment continued only 
six days. Observations were made in both experiments of the body 
functions, body measurements, pulse-rate, urine, feces, and respiratory 
exchange, and the computations and conclusions are of fundamental 
importance. They will be continually referred to in connection with 
this report. 

Observations by van Hoogenhuyze and Verploegh. 3 — In a study made of 
the urine excreted by a professional fasting woman, van Hoogenhuyze 
and Verploegh gave especial attention to the creatinine content. The 
experiment began on June 11, 1905, and ended June 25, 1905; the 

iPaton and Stockman, Proc. Royal Soc. of Edinburgh, 1888-1889, 16, p. 121. 
"Lehmann, Mueller, Munk, Senator, and Zuntz, Archiv f. path. Anat. u. Physiol, u. f. klin. 
Med., 1893, 131, Supp., p. 1. 

3 Van Hoogenhuyze and Verploegh, Zeitschr. f. physiol. Chem., 1905, 46, p. 415. 



PREVIOUS OBSERVATIONS OF PROLONGED FASTS. 15 

following constituents of the urine were determined: Total nitrogen, 
urea, creatinine, uric acid, chlorides, phosphoric acid, indigo, and total 
acidity. 

Observations by Brugsch, Mohr, Bonniger, Baumstark, and Hirsch. 1 — 
An experiment made on a fasting woman by Brugsch, Mohr, Bonniger, 
Baumstark, and Hirsch was continued from March 10 to March 25, 
1906. The observations included loss in body-weight, total nitrogen, 
and especially acetone in the breath and acetone and /3-oxybutyric acid 
in the urine. The ammonia-nitrogen was likewise determined. The 
research is of peculiar importance in that special emphasis was laid 
upon the relationship between acidosis and fasting. 

Observations by Cathcart. 2 — An experiment, carried out by E. P. 
Cathcart on the professional faster Victor Beaute, with the strictest 
surveillance, was designed primarily to study the effect of fasting upon 
the partition of the nitrogen, the new methods introduced by Folin 
being used. The experiment was made in Glasgow in 1907 and con- 
tinued 14 days. An especial study was made of the mineral matters 
excreted and the creatine and creatinine content of the fasting urine. 
An interesting complement to the fasting experiment was a study made 
at the end of the effect of the ingestion of the starch-cream diet of 
Folin, i. e., a low nitrogenous diet, which was continued for a few days. 
During this time the uric acid and purine-nitrogen were accurately 
determined and the chlorine, phosphorus, and the several forms of 
sulphur were carefully estimated. This investigation represents the 
most comprehensive and exact observation of the constituents of urine 
passed while fasting to be found in the literature. Cathcart's associate, 
Charteris, published his blood findings somewhat later. 3 

Observations on Gayer. — An uncontrolled fast of 30 days was made in 
New York on a professional faster, Gayer, continuing from May 16 to 
June 14, 1910. Although the attending physicians are by no means 
unanimous in their opinions regarding the genuineness of the fast, the 
body-weights reported in a non-scientific publication 4 indicate a loss 
in weight not unlike that experienced in accredited fasting experiments. 
The accurate blood examination made by Dr. I. S. Wile 6 inspires confi- 
dence in the report of this fast. 

Observations by Grafe. — Although complicated by abnormal psychical 
conditions, by an error on the part of the nurse in giving a rectal enema 
on the seventh day, and by considerable variations in muscular activity, 

brugsch and Hirsch, Zeitschr. f. exp. Path. u. Therapie, 1906, 3, p. 638; Bonniger and Mohr, 
ibid., p. 675; Baumstark and Mohr., ibid., p. 687. 

"Cathcart, Biochem. Zeitschr., 1907, 6, p. 109; Journ. Physiol., 1907, 35, p. 500; Cathcart 
and Fawsitt, Journ. Physiol., 1907, 36, p. 27. 

'Charteris, Lancet, 1907, 173, p. 685. 

*Long, Physical Culture, August 1910, p. 190. 

'Personal letter received from Dr. I. S. Wile, dated December 24, 1912. 



16 A STUDY OF PROLONGED FASTING. 

the experiment of Grafe 1 with the Jaquet respiration apparatus 2 at the 
Medical Clinic in Heidelberg is of interest in throwing light upon the 
gaseous exchange and the character of the katabolism during prolonged 
inanition and on the ratio of carbon to nitrogen in fasting urine. Fur- 
thermore, it substantiated the observations made by Brugsch and 
others on the acidosis during fasting, as indicated by the excretion of 
acetone and ,8-oxybutyric acid. 

Observations at Wesleyan University, Middletown, Connecticut. — With 
a special view to studying the drafts upon body-material during fasts 
of 24 to 168 hours, a lengthy series of experiments was undertaken at 
Wesleyan University, Middletown, Connecticut, the results of which 
have already been published. 3 These experiments threw much light 
upon the character of the drafts upon body-material during the experi- 
mental periods and showed that the organism and particularly the 
storage of glycogen in the body may be greatly affected by even a 
short fast. It has furthermore been shown that glycogen — the body- 
material which is first and most heavily drawn upon during fasting — may 
be considered as one of the most quickly realizable assets, the removal 
of which affects profoundly one of the safety factors of the human body. 

OBSERVATIONS ON SUCCI. 

Fast in Florence, 1888. — Although a short account of the experiment 
on Cetti made by Lehmann, Mueller, Munk, Senator, and Zuntz was 
published in 1887, 4 the details of their investigation did not appear 
until 1893, 5 and the first extensive report of a prolonged fasting investi- 
gation was that made by Luciani of the fasting experiment with Succi 
in Florence during the spring of 1888. The Italian report of this fast 
was published in 1889, 6 but the work is best known to other than 
Italian readers by Fraenkel's translation. 7 

Luciani's study of Succi included an extensive series of observations. 
Unfortunately, since the partition of the nitrogen in the urine was at 
that time imperfectly understood and as the gaseous exchange was 
studied under conditions affecting seriously the accuracy of the results, 
Luciani's observations are more especially of value as indications of 
the general body functions of a fasting man than as measurements of 

^rafe, Zeitschr. f. physiol. Chem., 1910, 65, p. 21. 

2 Geheimrat W. His, on a recent visit to the Nutrition Laboratory, informed us that a fasting 
experiment with a professional faster, a woman, continuing 4 weeks, had been carried out not 
long before in his clinic in Berlin by Professor Staehelin, in which the Jaquet respiration apparatus 
had been used. Owing to the indisposition of the subject, the experiment was less successful 
than had been hoped, and Professor Staehelin's removal to Basel has indefinitely postponed the 
publication of the results. 

"Benedict, Carnegie Inst. Wash. Pub. No. 77, 1907. 

4 Lehmann, Mueller, Munk, Senator, and Zuntz, Berliner klin. Woch., 1887, pp. 290 and 425. 

"Lehmann, Mueller, Munk, Senator, and Zuntz, Archiv f. path. Anat. u. Physiol, u. f. klin. 
Med., 1893, 131, Supp., p. 1. 

•Luciani, Fisiologia del digiuno; studi sull' uomo. Florence, 1839. 

7 Luciani, Das Hungern. Translation by M. C. Fraenkel. Hamburg and Leipsic, 1890. 



PREVIOUS OBSERVATIONS OF PROLONGED FASTS. 17 

specific chemical transformations. Succi's peculiar psychical condi- 
tion, a condition which seems to be characteristic of the ascetic who 
subjects himself to a fast of 30 days or more, is interestingly commented 
upon in extenso by Luciani. The research as a whole was a model in 
plan, and as a painstaking record of cooperative research in fasting it is 
equaled only by the experiments of the Berlin investigators. Luciani's 
study unquestionably stimulated the considerable number of experi- 
ments subsequently carried out with Succi, at least 7 experiments, each 
continuing 20 or more days, being made with him by different investi- 
gators and in different places. 

Fasts in Milan and Paris, 1886. — In reporting the results of the 
Florence fast, Luciani refers to two fasts said by Succi to have been 
made previously, one in Milan in August and September 1886, and a 
second in Paris in the latter part of November and the early part of 
December 1886. The short time between the fasts is of special interest. 
Little is known regarding these two fasts, but Luciani considered the 
records of the body- weights obtained from Succi' s notebooks sufficiently 
reliable to include in the published report of his research and he plotted 
curves from them showing the loss in body- weight during the fasts. 

Fast in London, 1890. — In 1890 Succi carried out a 40-day fast in 
London, which began on March 17. 1 Although observations were 
made of a number of factors during this fast, the controls were so 
incomplete that, aside from the body-weight, the observations have but 
little value at the present time. The body-weights were apparently 
recorded with a great degree of accuracy and form the basis of a curve 
which will be discussed later. No statements accompanied the records 
of the pulse and respiration as to whether the subject was lying, sitting, 
or standing, so that they can have but little significance; fluctuations 
in the pulse-rate give evidence of marked changes in the muscular 
activity at times. Strength tests were made with a hand dynamometer 
each day, showing practically no alteration in the strength. The 
axillary records of the body-temperature indicate a lowering of the 
temperature toward the end of the fast. 

Fast in New York, 1890. — According to Succi's own statements, 
substantiated by newspaper reports, Succi carried out a large number 
of fasts which were not scientifically controlled. One of the most 
important of these was made in New York City about 8 months after 
the London fast. 2 The New York fast began on November 6, 1890, 
and was said to have continued 45 days. Correspondence with several 
of the physicians who attended this fast shows a diversity of opinion as 
to its authenticity. On the other hand, the body-weights recorded, if 
correct, indicate about the usual loss in weight, the records being 147.4 

British Med. Joum., 1890, pp. 764, 819, 876, 935, 996, and 1056, also p. 1444. 
2 The New York Daily Tribune, November 6, 1890, and December 21, 1890. 



18 A STUDY OF PROLONGED FASTING. 

pounds (66.86 kilograms) at the beginning of the fast and 104.75 pounds 
(47.52 kilograms) at the end. 

Fast in Naples, 1892. — The next scientifically controlled fast with 
Succi was in Naples, beginning August 7, 1892. Observations were 
made by Ajello and Solaro, 1 most of these being on the urine. The 
body-weight was likewise carefully recorded as the fast progressed, as 
well as the amounts of water taken. The determinations made on the 
urine which are of interest at this time are those of the chlorine and 
phosphoric and sulphuric acids. On the second day of the fast, 2 
grams of feces were passed and on the eleventh day, 317 grams. 

Fast in Rome, 1893. — A number of observations were made on Succi 
by Dutto and Lo-Monaco 2 during a 20-day fast in Rome beginning 
December 16, 1893. The body- weight was recorded each day, also 
the amount of water taken. Analyses were made of the urine excreted, 
these being much more complete than in any of the earlier fasts, as the 
nitrogen was determined by the Kjeldahl method. Determinations 
were also made of the acidity of the urine and the content of sulphur, 
ethereal sulphates, neutral sulphur, chlorine, phosphorus, sodium, and 
potassium. 

Fast in Vienna, 1896. — The urine excreted by Succi in a 21-day fast 
was studied by E. and 0. Freund 3 in Vienna in 1896, an extensive 
partition of the nitrogen being attempted for the first time. The 
observations as to Succi' s condition, including the body- weight, were 
unfortunately lost. 

Fast in Zurich, 1896. — During a 21-day fast of Succi in Zurich, 
beginning September 13, 1896, Daiber 4 studied the urine and obtained 
the body-weight. The body-temperature, the amount of water taken, 
and the chlorides in the urine were all determined with sufficient accu- 
racy to make them of value at the present day. 

Fast in Hamburg, 1904-. — The last recorded experiment on Succi was 
made in Hamburg in March 1904. During the last 10 days of this 
30-day fast, the urine was examined by Brugsch, 6 who determined the 
partition of the nitrogen. Special emphasis was laid upon the acidosis. 

^Ajello and Solaro, La Riforma Medica, 1893, 2, p. 542. 

2 Dutto and Lo-Monaco, Policlinico, 1895, 2, p. 1. 

3 E. and O. Freund, Wiener klin. Rundschau, 1901, 15, pp. 69 and 91. 

♦Daiber. Schweiz. Woch. f. Chem. u. Pharm., 1896, 34, p. 395. 

5 Brugsch, Zeitschr. f. exp. Path. u. Therapie, 1905, 1, p. 419. 



FASTING 



PLATE 2 




C. Respiration experiment made by T. M. Carpenter, on the Universal Respiration Apparatus. 
These experiments were made each morning, just after the Subject left the Respiration Calori- 
meter, and before he stood up. 




D. Weighing the Subject on the Thirty-first day of the Fast. At the right is shown the Bed on 
which he has just finished the Respiration Experiment ; the Universal Respiration Apparatus 
is shown at the extreme right. 



RESEARCH ON METABOLISM IN PROLONGED FASTING AT 
THE NUTRITION LABORATORY. 

PROBLEMS TO BE STUDIED. 

In the research on metabolism during short fasting periods, which 
was carried out at Wesleyan University, Middletown, Connecticut, the 
changes incidental to the first days of fasting were, it is believed, ade- 
quately studied. On the other hand, it was desirable to supplement the 
earlier observations by a study of the metabolism during prolonged 
fasting, since many points regarding the course of the metabolism 
after the body had adjusted itself to the fasting condition had not been 
established. For instance, as the fast progresses it is important to 
know whether the gross metabolism alters either per kilogram of body- 
weight or per square meter of body-surface, also whether the acidosis 
is extreme or whether there is an acquired tolerance of it, and what 
effect the acidosis, if present, has upon the metabolism. Since the 
carbon, the ammonia, and the heat of combustion of the urine, also 
the composition of the alveolar air, give indications as to acidosis, a 
study of prolonged fasting should include determinations of all of these 
factors. In the earlier fasting study determinations were made of a 
number of the constituents of the urine, including total solids, nitrogen, 
creatine and creatinine, phosphorus, sulphur, and chlorine. In the 
longer research it would be necessary to elaborate these determina- 
tions, studying also the composition of the feces, should any be passed 
during the period. Furthermore, the relationship between the pulse- 
rate and the metabolism, the character of the respiration as shown by 
graphic records, the variations in the body-temperature, and the changes 
in the composition of the blood, all have sufficient significance to 
warrant investigation. Since muscular activity has so great an influ- 
ence upon metabolism, the experiments of Zuntz on Breithaupt should 
be duplicated with more modern technique. Comparison should be 
made of the metabolism in selected periods with constant external 
conditions instead of with changing activity as in the earlier research, 
and experiments in which the subject breathed a high oxygen atmos- 
phere would also be desirable. 

The Nutrition Laboratory was especially fitted to carry out a research 
of this kind, being well equipped with apparatus for determining the 
respiratory exchange and the heat output, as well as for measuring the 
pulse, respiration, and muscular activity. It was therefore of funda- 
mental importance to have ready a carefully prepared plan for studying 
the metabolism during prolonged fasting which could be used when- 
ever an opportunity offered for conducting such a research. On the 
other hand, it was not desirable to make undue haste in beginning the 

19 



20 A STUDY OF PROLONGED FASTING. 

study, inasmuch as the equipment of the Laboratory was steadily being 
increased. The chemical technique was also being rapidly perfected, 
the development of the new micro methods of Professor Folin being of 
especial value in studying the relatively small volumes of urine excreted 
during prolonged fasting. 

SELECTION OF SUBJECT. 

While no particular effort was made to secure a subject for this 
research, advantage was taken of a visit to New York by Succi to confer 
with him. His age and his somewhat unreasonable demands for a 
large compensation made an arrangement with him undesirable. Fur- 
thermore, he would not have cooperated readily in the great number of 
tests that were included in the plan for the fasting research. A number 
of individuals, stimulated by the report of the earlier study, offered 
themselves to the Nutrition Laboratory as subjects for a fasting experi- 
ment. A large majority of these were either sufferers or imagined that 
they were sufferers from "nervous disease," and were therefore patho- 
logically or psychologically undesirable. Furthermore, none of them 
had a clear conception of a scientifically controlled fast and of the 
importance of the observations which would be included in such a 
research. They were therefore not seriously considered. 

In the spring of 1911, a letter was received from A. Levanzin of Malta, 
offering himself as a subject for a long fasting experiment to be carried 
out at the Nutrition Laboratory. The letter was voluminous, but very 
intelligently written, and showed an appreciation of the scientific value 
of such a research. As Professor Luciani, of Rome, who had made 
the classical study with Succi, later expressed his confidence in the 
ability of Levanzin to carry out a fast of this length, it seemed probable 
that the subject desired for the research had been found. It was sub- 
sequently learned that Professor Luciani's acquaintance with Levanzin 
was through correspondence only, but his recommendation went far 
to convince us of the desirability of attempting an experiment with 
this man. Accordingly an exact statement was sent A. L. of the duties 
involved in a research of this nature and an arrangement was entered 
into for him to come to Boston for the purpose. In accordance with 
his own proposition, the agreement was made to cover his expenses, with 
a bonus if the experiment was successfully completed, and every 
attempt was made to minimize anxiety on the part of the subject. The 
risk of protracted illness incidental to the journey from Malta to Boston, 
to the change in climate, and possibly as a result of the fasting experi- 
ment, had to be considered, and a sworn statement exonerating the 
Nutrition Laboratory from any responsibility for illness of more than 
4 days' duration was obtained from L. before he left Malta. 



PROOFS OF PHYSICAL FITNESS. 



21 



PROOFS OF PHYSICAL FITNESS. 

It was necessary to assure us as far as possible of the fitness of this 
man for the research, and he was requested to send us a physician's 
certificate as to his health. These proofs were supplied and were as 
follows: 

Ratnapoora, Sliema, Malta, 10th January, 1912. 
I hereby certify that Mr. A. Levanzin, B. A., is in good health. He does not 
suffer from any disease and his organs are healthy. 

(Signed) Robt. Samut, 
Professor of Physiology of Malta University. 

Examination of Urine submitted by Mr. Levanzin on Feb. 10, 1912. 



Quantity in 24 hours: Unknown; taken as 1500 c.c. 



Color: 


Light amber. 


Odor: 


Sui generis. 


Reaction : 


Acid. 


Specific gravity: 


1018. 


Total solids: 


41.9 


Deposit : 


None. 


Urea: 


2.1 per cent. 


Uric acid: 


0.03 per cent. 


Chlorides: 


1 per cent. 


Phosphates: 


0.35 per cent. 


Indican : 


Nil. 



Microscopical examination: Negative. 



(Signed) 



Abnormal constituents. 

Albumin 

Peptone 

Globulin 

Glucose 

Acetone 

Blood r None. 

Bile 

Pus 

Mucus 

Diazo reaction 



Robt. Samut, Edinb. 



Roseville, 39 Strada Ghar-id-dud, Sliema, Malta, 

January 20, 1912. 
This is to certify that I have to-day physically examined Mr. Agostino 
Levanzin and that I have found him in good health and free from organic 
disease. 

(Signed) Jos. S. Galigia, M. D. 

52 Victoria Terrace, Sliema, Malta, November 7, 1911. 

I hereby certify to have examined A. Levanzin, Esq., B. A., Ph. Ch., P. L., 
and have found him in a good state of health. The urine was normal in every 
respect. Specific gravity, 1025. No traces of albumen nor those of glucose, 
etc., have been detected. His height is 5 feet 6 inches. His skeleton and 
muscles are normally developed. His gross weight is 152 lbs. I am of 
opinion that he could undergo quite easily under ordinary circumstances a 
period of prolonged fasting without detriment or danger to his health, and that 
under ordinary conditions he is not liable to suffer from any illness that might 
upset the experiment or entail any hindrance to same. 

I know Mr. Levanzin since many years and in fact I am his family doctor. 
I might add that he has already fasted for a long period without suffering any 
serious bad after-effects; indeed, I was astonished at his rapid recovery there- 
from and return to his normal state of health. * * * 

(Signed) Dr. P. P. Agius, B. A., Ph. Ch., M. D. 



22 A STUDY OF PROLONGED FASTING. 

AUTOBIOGRAPHICAL NOTES. 

On the twenty-ninth, thirtieth, and thirty-first days of his fast at the 
Nutrition Laboratory, L. wrote a sketch of his life. This is reproduced 
verbatim, since it shows many of the interesting features of the life, 
education, and habits of thought of the subject. 

12th of May, 19 1 2 (29th day of my fast). 

More than one hundred years ago, Gabriele Avanzino, a Sicilian, settled in 
Malta. Gradually the surname was corrupted into Levanzin. My mother, 
Lorenza Borg, living and aged about 58, descends from pure and noble Maltese 
blood since 400 years. Her grandfather's uncle was the famous Vincenzo 
Barbara, the daring sea-captain of one of the French battle-ships who was by 
Botta and other historians falsely accused of having betrayed Marat when he 
landed him to take possession of Naples on behalf of Napoleon. Barbara was 
the right arm of Napoleon to plot and get rid Malta from the yoke of the 
Knights of St. John and he was also the first Grand-Master of Free-Masons 
in the Island. Her grandfather was Joseph Borg, another sea-captain who 
came to America in the time of the Revolution, volunteered with the insurgents 
and fought for the American independence many battles as in his portrait that 
we keep he has on his breast from seven to eight medals. That is why I 
probably love so much freedom, independence of thought, and sympathize 
keenly with America. 

My father, Paolo, living and aged about 68, is also the son of a sea-captain, 
Agostino, who was drowned when my father was only 3 years of age and so 
could not nave a liberal education. He learned the art of ship-building which 
was very flourishing in those commercial times, but now being disabled from 
both his hands through two accidents that happened to him during his work, 
he is carrying a grocery-shop in a village as my mother is carrying a confec- 
tionery and toy one in the same place. They are both very honest and hard- 
working people and although they have sufficient property to keep them up 
comfortably during their old days, they do not want to give up their business 
as they want "to leave us something after their death." I have a sister, 
Teresina, 20 years, living with my mother and a married one to an engineer, 
Ursola, 28 years. 

I was born in the Citta Cospicus of Malta, on the 23rd of May, 1872 — 40 
years ago. At 6 years of age I went to Egypt with my mother where my 
father was working but came back after two years as the hot climate did not 
suit us. Frequented the public free-schools and at ten had my first prize — a 
five shilling piece — for writing the best essay against "Cruelty to Animals." 
Then prizes for drawing as I am very fond of art especially of music and 
painting. At 12 I entered the free Dockyard Schools and had several prizes. 
At 14 I was admitted by competitive examination as shipwright apprentice as 
I wished to follow my father's career, then promoted to draughtsman and then 
to clerk. From infancy I was always inclined to hard study and sometimes 
during the night I used to steal out of bed to read some interesting book 
because my parents did not like to see me overstrain my already weak eyes. 

During the time that I served my apprenticeship in the Dockyard I published 
two weekly papers, successively, in Maltese the "Habil ta Cullhadd" (The 
Friend of All) and " Is-Sengha" (Art) to educate and enlighten the working 
classes that live in a very miserable condition and are totally forsaken by the 
Government, but both papers failed after a few months through lack of sub- 
scribers. At 17 I felt inclined to follow the ecclesiastical career to devote 



AUTOBIOGRAPHICAL NOTES. 23 

myself entirely to study and oratory, that I like so much, and became a cleric, 
but after four years, through matter of convictions and bigoted tyranny of the 
superiors, I put off my black robe and entered the Lyceum to prepare myself 
for a professional career. 

At 20 (1892) I passed my matriculation examination and took up the medical 
courses. At the same time I was contributing literary and political contribu- 
tions to our best papers and published several poems in Italian that were very 
favorably appreciated by the press. I started also the publication of a 
University Magazine "Lo Studente Maltese" to stimulate the other students to 
contribute literary and scientific articles and I published in English and 
Italian a study on Shakespearean drama and some biographies of eminent 
Maltese personages. The paper dragged a stinty existence for two years and 
perished through lack of funds. At the same time I was conducting two other 
political papers in vernacular (Maltese), the " Cottonera" and the "Habil ta'l 
Poplu," and it was one of the articles contributed to the "Cottonera" that 
provoked against me my first libel and was tried by jury. 

My father was still working in Dockyard and as his foreman used to take 
bribes from his employees and borrow from them money that he never used to 
return back, and as my father did never like to satisfy him in this because he 
fulfilled always all his duties honestly and regularly he became his scapegoat 
and was always ordered to do the most dangerous and hard kind of work. 
Twice he was hurt, twice amputations had been operated on fingers of both 
hands, with peril to his life, till he became a disabled man. I protested to the 
superiors and they answered that they did not care a bit about it and so, at 
last, I published in the "Cottonera," in 1895, a violent article in English in 
which I enumerated with details the many bribes and irregularities that were 
continually committed in H. M. Dockyard, signed the article and defied the 
Admiral Superintendent that I was ready to prove in court all my assertions. 
The article provoked a great scandal and the Admiral was obliged to arraign 
me before the criminal courts to prove my assertions. The penalty demanded 
against me was six months of hard labor imprisonment and a fine of £500. 
All my assertions were proved to the very hilt after a fierce fight and I was 
triumphantly acquitted, unanimously, by the jury. As I was defending the 
cause of thousands of leech-bled victims against a few vampires I was triumph- 
antly carried on the shoulders of the workmen, with bengala-fires and bands 
playing, but the next morning my father was discharged from the Dockyard 
and lost his bread that was keeping us!!! I felt the shock tremendously but 
did not discourage myself. I put myself in correspondence with Mr. Labou- 
chere of the "Truth" of London, who not only published my contributions in 
his very influential paper but brought the matter before Parliament, being an 
M. P., and fought it out very bravely. A Commission was sent to Malta 
and all my statements have been found to be true, my father was put to work 
again, and several important reforms were introduced. But after a few 
months my father was discharged again and forever!!! under the free and 
glorious banner of liberal Britain!!! 

The libel took place on the 7th of August, 1895. In September of the same 
year, I took my degree of Bachelor of Arts from the Malta University after 
obtaining for three years a 50 per cent in higher mathematics, physics, natural 
history, philosophy, Latin, English and Italian literature and history. But 
my father about that time was out of work and so I had to add to my already 
overstraining work private lessons after my lectures, sometimes till 10 p. m., 
and plodded on in this very hard and anxious life for about two years in which 
I have followed successfully the Anatomical, General and Pathological, the 
Dissectional, the Physiological, the Obstetrical, the Surgical, the General 



24 A STUDY OF PROLONGED FASTING. 

Pathology, the Chemistry, the Bacteriology, the Materia Medica, the Thera- 
peutical and the Pharmaceutical Courses. But as at that time I was under the 
false impression that as I was working mentally very hard I had to eat more 
and more, I used to stuff myself with a lot of meat and eggs and milk and these, 
added to the great overstrain, shattered my nervous system down with a severe 
shock of neurasthenia. My professors gave me the good advice to take a long 
rest and to suspend my studies for a prolonged period of time. But my family 
could not afford that for my father was not working all the time and I had to 
work to live. So I took the warrant as a Pharmaceutical-Chemist after a 
severe examination and was employed as director of the most important 
pharmacy in Valletta (the capital of Malta), called "Mizzi's Dispensary." 
I lived there for a year and my neurasthenia got a little better through enforced 
rest. 

But I was living away from my family and had to run into many expenses 
to have my meals in hotels and I was always sleeping in the pharmacy not to 
cross the sea late in the night and go home. So I employed myself in a phar- 
macy at Cospicua, very near home, and lived there for about two years. My 
father and mother at the same time started their business and were progressing 
very prosperously. My wife, Lucia, lived just opposite, and we loved each 
other. I married her on the 24th of April, 1900. She is the eldest daughter 
of Doctor G. F. Inglott, Medical Officer to Government, Knight of the Pope, 
and member of several literary and scientific academies and is considered as 
the most clever obstetrician and gynecologist in Malta, enjoying a very wide 
practice. So I was determined by him to start a pharmacy of my own, which 
I did and the result was a very successful one, but a short time after the 
Transvaal War broke out and as he is well conversant with the English lan- 
guage was called by the military authorities in charge of the Military Hospital 
and so all his time was absorbed in these exacting duties and could not take 
care any more of his private practice. This lasted for over two years and at 
last the pharmacy broke down and I had to remove to a wealthy country 
district called Birchircara. 

18th of May, 1912 (80th day of my fast). 

Before going to live in Birchircara I had fought two great battles — one on 
behalf of down-trodden and neglected Democracy and the other one advocating 
the maintenance in our tribunals of the Italian language that has been the 
means of our civilization since about 600 years. I have founded the first 
"Malta Trade Union, ,, of which I was elected President, with 700 members, 
free schools, lectures, honest amusements, band, and carried it on successfully 
for some time, but political intrigue not to encourage a labor party and not to 
enlighten the lower class made it dwindle into nothingness and all my " love's 
labor was lost." 

Then I went to Italy, at my own expense, for about a month, to lecture 
against Mr. Chamberlain's (England's Prime Minister at that time) edict that 
the Italian language had to be cleared off from our courts within a lapse of 
fifteen years. The movement had some good effect, because all the Italian 
press was awakened and protested loudly and vigorously and Mr. Chamberlain 
had to give up his Order in Council. 

After creating in Birchircara a prosperous practice for my pharmacy I 
wished to provide for the future, and as my neurasthenia was progressing 
through the very close and sedentary life that I was conducting, shut up from 
7 a. m. to 10 p. m., including holidays, I determined to secure a more easy 
career — law. I entered the legal course and succeeded to obtain a warrant. 
But to continue to carry on the pharmacy, to keep up my family, and to follow 



AUTOBIOGRAPHICAL NOTES. 25 

a difficult university course was a very severe test on my already shattered 
nerves, and always under the false idea that to work very hard I had to overeat 
and to stuff myself with as much protein as possible, I ruined my health to such 
an extent that I was compelled to give up my pharmacy forever and dedicate 
myself to the practice of law that offered more leisure and also better prospects 
for me as I was and am still very popular and beloved by the people. Fortu- 
nately enough to help me at the start of my legal career, I was offered at Sliema 
(a beautiful summer resort in Malta) the management of a pharmacy with a 
very good salary with the permission to absent myself during the morning 
hours to go to court and plead my cases. So I went to live there and Miranda 
Cordelia was born, while Jolanda Beatrice was born in Birchircara. My legal 
practice prospered so rapidly that after a year I had to give up my pharmacy 
management and dedicate myself entirely to the legal career that I am still 
following at present. 

When I thought to have fixed a solid basis for my family's subsistence I tried 
again to do some good work for the cause of our trampled down and utterly 
neglected lower classes. It has been always my ideal to enlighten them, to 
help them to push themselves forward as the workmen of other more progres- 
sive countries do, because although I have parted from their class my demo- 
cratic soul was always with them. So I started the publication of a weekly 
paper entitled "In Nahla" (The Bee) the scope of which was to instruct 
in scientific, artistic, historical, and literary knowledge, as plainly and as enter- 
tainingly as possible. The effort was a brilliant success because I had immedi- 
ately the greatest circulation ever attained by any paper published in any 
language in Malta. My wife cooperated herself very effectively because she 
contributed, every week, some interesting article about the rearing up of 
babies, hygiene, against the marriage of consumptives or between relatives, etc. 

I have published in the same paper a historical novel "Is Sahhar Falzon" 
(The Wizard Falzon) in which I have treated fully and faithfully all the 
history of the first 60 years of the dominion of the Knights of Malta over the 
Island from Lisleadam to La Cassiere. My intent was to teach to the people 
its history not in the usual pedantic and monotonous way but enhancing it by 
intermingling to it the attractive episodes of chivalry and love. In the third 
part of the novel I have tried the scientific novel trying to popularize science in 
a delectable and easy way as I have done with history, and as Falzon was a 
Roman Catholic priest who was burned up alive accused of witchcraft, I 
developed all the up-to-date positive knowledge about psychical science of 
which I am an ardent and keen student. In many notes I have suggested the 
best books and authors and described the most authoritative experiment for 
those who wished to delve deeper into the matter. All the facts about Falzon 
were gathered through a lot of poking in our archives amongst very rare 
manuscripts of those past, dark, and barbarous ages. This novel was a great 
success because I had to publish separately in three volumes comprising over 
650 large pages and the edition was sold out very rapidly. 

In the "Nahla" I have not only tried to instruct the lower classes but I have 
fought hard also to defend their rights and to uplift my voice for the injustices 
committed against them. Twice I was tried by jury for libelous articles but 
twice I was triumphantly acquitted. The first time was on the 28th of 
October, 1909, when Antonia Azzopardi, a murderer, was hanged. The 
doctors in charge had executed their post-mortem examination so carelessly 
that there was doubt that the man was buried alive only one hour after the 
execution! I accused them of that in a very violent article, and all Malta 
was in a devilish row about it. The Governor ordered the Chief Medical 
Officer, who was responsible, to libel me and after a very hard struggle before 



26 A STUDY OP PROLONGED FASTING. 

the jury I have succeeded to prove that there were no positive and scientific 
facts to prove that the executed man was dead when buried. This result 
provoked a new law in Malta and now instead of burying the executed men 
after only one hour from the execution as before, they watch them keenly for 
24 hours, and as I protested also that it was barbarous to bury them in a sack 
after that Justice had made its cold vengeance on a creature of God against 
whose life she has no right at all, now they bury them in a cheap coffin. 

The second trial was provoked by this fact. To communicate by means of 
telephone in Malta you have to pay 60 cents, and the telephones are at the 
Police Station. Poor people are supplied gratuitously by Government with 
doctors, midwives, and medicines. At a village called Zeitum a very poor 
woman was dying through post-partum hemorrhage. The midwife sent for 
the doctor for assistance as she thought the case a fatal one. The doctor 
happened to be in another village, and the policeman refused to call him 
immediately before levying the tax of the telephone. The poor woman had 
not the 60 cents to pay for it, and more than an hour was spent till they got 
them from a distant sister. When the doctor arrived there was no more hopes 
to save her and the poor victim of human brutality died leaving a husband and 
six orphans. I published a violent attack against the police accusing them of 
manslaughter and was libelled, but having proved to the hilt all the facts 
stated, I was again acquitted triumphantly by the jury. 

As you can see my "Bee" was really a "busy" one and played very well and 
smartly her humanitarian and democratic mission. At the same time we did 
not miss to advocate, and very ardently, " Fletcherism " and the Fasting Cure 
for the cure of disease as also many other important dietetic reforms. Many 
articles were also published on behalf of the idea of an international Language. 
A lecture in Italian that I delivered in Malta several years ago advocating 
Esperanto was published in it. About 25 years ago I learned Schleyer's 
" Volapuk" that broke down, substituted by "Idiom Neutral," a more national 
system. I follow my friend Rosenberger of St. Petersburg and learned it also 
but had very little success. Then my dear friend's Dr. Zamenhof of Warsaw 
"Esperanto" came in vogue and I learned it and took up arms in favor of it 
very ardently. I have given in Malta free courses in the University, lectures, 
founded societies and succeeded also to start the first female course in the 
University in any branch of knowledge. Mrs. Levanzin was a great help to 
me in this movement and now she is the first woman in Malta to enter the 
University to follow a medical career. She is trying with all her efforts not 
only to enlighten the female classes of Malta that are yet shrouded in mediaeval 
darkness by publishing very instructive articles but also by setting them the 
good example of opening for them new and prosperous careers. Esperanto 
had a great vogue in Malta; I, with Mrs. Levanzin, took part in the Inter- 
national Congress of Barcelona and there I was elected "President of the 
International Association of Pharmaceutical Esperantists," editor of the 
scientific Esperanto monthly, "La Vocho de Farmacustoj," and Corresponding 
Member of the "Colegio des Farmaceuticos " (the oldest one in the world and 
where the first pharmacopaeia was published) after my lecture in Esperanto on 
the "Fungus Melitensis" by colleagues of over 30 different nationalities. 
Now I am advocating "Ido" or Simplified Esperanto as I find that it is easier, 
more logical, and cropped of all the errors and incongruities contained in Dr. 
Zamenhof's system. 

14th of May, 1912 (31st and last day of my fast). 
I have also at the same time fought hard against much ridicule and prejudice 
to found the first "Society of Psychical Studies and Research" in Malta of 
which I am President. Honorary Members are Prof. Crookes, Russell 



AUTOBIOGRAPHICAL NOTES. 27 

Wallace, Lodge, Maxwell, Richet, Lombroso, Morselli, Carrington, etc. Now 
another battle for Science and Humanity — Fasting. About two and a half 
years ago, while I was over-eating, obese, neurasthenic, pessimistic and with a 
shattered nervous system, I chanced to read in the "Contemporary Review" an 
article about fasting. It was a flash of light that struck me vividly. It indicated 
to me the right path to health and happiness and I followed immediately its 
dictates with enthusiasm. I fasted for 8 days with very great benefit. Then 
I procured all the possible literature in several languages about fasting and 
prepared myself thoroughly for a whole year for a long and "conquest" fast. 
I started that on the 1st of March, 1911, and Mrs. Levanzin did the same as 
she had been suffering since several years from severe dyspepsia and insomnia 
through over-eating. She broke her fast on the 33rd day and I on the 40th 
with immense benefit to our health because our ailments disappeared. We 
continued all our usual occupations during our fast and did never feel any bad 
effects. 

In the following August, cholera broke out in Malta and as a preventive 
precaution I fasted again for 12, Mrs. Levanzin for 17, and my daughters for 
several days each. I have cured Jolanda from a severe case of small-pox by 
17 days of fasting and Miranda from a severe case of fever with 8 days. 
Several other friends and parents underwent the cure of fasting under my advice 
with marvelous effects. Enthused by these beneficial results, I determined to 
fix a scientific basis to it by undergoing a thorough and seriously controlled 
experiment under the direction of a physiologist of high repute and great 
experience. I submitted the case to my friend, Professor Luciani, of Physi- 
ology, of Rome, who studied Succi and published a good book on the "Physi- 
ology of Fasting," and he suggested to me to come over to Boston at the 
Carnegie Institution, * * * as the Institution was the best equipped in 
the world for such an important experiment. I took up his suggestion and 
crossed over 5,000 miles to undergo my fast, refusing any pecuniary remunera- 
tion, only the expenses being defrayed for it. To-day is the 31st day and last 
day of it, and I can simply tell you that it is a complete success. I am feeling 
very well, very uplifted, and I wished to prolong it further, at least to 40 days, 
because I do not feel yet any trace of hunger at all. But Professor Benedict 
thought it already very expensive and fatiguing and bid me to break it to- 
morrow. He only allowed me to prolong it for a day more, simply to beat the 
record of the longest controlled scientific fast ever made. During the fast I did 
not feel the least uncomfortable sensation except the bad taste of my coated 
tongue, and the catarrh and congestion of my eyes that I had at the start 
have nearly disappeared. I hope that a great benefit to my health shall 
accrue from it. 



GENERAL CHARACTERISTICS OF SUBJECT. 

As will be inferred from his biographical notes, L. was a propagandist 
with pronounced views on all subjects. He had had some legal training 
and was inclined to be exceedingly contentious. His chirography was 
excellent. He also had a good command of the English language, as well 
as of Italian, French, Spanish, Maltese, and Esperanto. His familiarity 
with the vagarious literature on fasting was astonishing, and led him 
to make many suggestions indicative of a mind working upon a propa- 
ganda for the supposed benefits to mankind to be derived from fasting, 
instead of an appreciation of the true scientific value of a prolonged 
fasting experiment. As an example of this, while he was unwilling to 
undergo a series of carefully planned strength tests, he nevertheless 
attempted some sensational strength tests of which he had read, such 
as lifting up a man and holding him suspended for a moment or two. 

He was a moderately well-nourished man, but his flesh was soft and 
flabby. This was natural, as he was decidedly sedentary in his habits 
and much averse to any muscular effort. It was hoped that measure- 
ments of the fasting metabolism during muscular work could be made 
with this subject by having him take a moderate amount of exercise 
daily on the bicycle ergometer, but he absolutely refused to mount the 
ergometer. He said he never rode the bicycle and thought it beneath 
his dignity, and that although the bicycle was used in Malta, it had not 
been employed by his people. As L. showed so strong an objection to 
muscular activity, we were obliged to omit these valuable observations. 

This subject called himself a vegetarian and frequently made a 
statement to that effect during the fast, but his practice did not wholly 
bear out his claim. He admitted that he ate meat in the European 
restaurants and on the boat during his trip to Boston, but said that it 
made him sick and uncomfortable, and that he was obliged to eat the 
meat, since he could not get the food he wished. Of considerable sig- 
nificance in this connection is his selection of food on the days preceding 
the fasting period. On his arrival in Boston he was taken to a hotel 
by one of the laboratory assistants and when given his choice of food 
from the menu, he ordered a large steak covered with onions; on other 
occasions he ordered salmon, pork, and lamb chops. During his stay 
in the hospital after the fasting experiment was over, he again called 
for a beefsteak. While probably not an excessive eater of meat, he 
was by no means a vegetarian for several weeks prior to the fast. 
The nitrogen found per day in the urine during the ten days preceding 
the fast indicated that he was living on a fairly high protein level. 
During the food days in Boston his diet was unrestricted and he was 
repeatedly told that he could have whatever he wished to eat during 
this preliminary period, except that it was preferred that the last meal 
of the day should not be excessively high in protein to avoid the long 
duration of the specific katabolic action of the protein. 

28 



GENERAL HISTORY OF FASTING EXPERIMENT. 

L. left Malta the latter part of February 1912, visiting Rome, 
Florence, Paris, and London, on his way to Liverpool. From the 
latter city he came direct to Boston, arriving at the Nutrition Labor- 
atory on the evening of April 10, 1912. Previous to his leaving for 
Boston, he had been asked to collect in 24-hour periods the urine 
passed during the trip across the ocean. While the conditions under 
which he would be living were necessarily abnormal, it was hoped by 
means of these specimens of urine to obtain some idea of the daily 
nitrogen outgo of the body. Specific instructions were given him 
as to the measurement, sampling, and preservation of the urine; as 
he had had a thorough training in pharmacy, he was well qualified to 
carry out the routine intelligently. The collection of the urine proved 
somewhat troublesome, as his roommates on the steamer could not 
appreciate the importance of the scientific test that he was to undergo. 

During his stay in the laboratory, he lived the entire time in the 
calorimeter room, except when he was taken out for a ride or to the 
roof for a change of air and scene. The calorimeter room is large and 
well-lighted and contains several calorimeters and respiration apparatus 
of various models, thus permitting a considerable number of observa- 
tions. The subject made his headquarters in a balcony of this room, 
but during the night he slept in a sealed calorimeter. 

The balcony in which he spent his time when he was not in the respi- 
ration chamber or on the respiration apparatus was supplied with a 
comfortable sofa, chair, and desk. A bottle containing a liter of distilled 
water was given him, also a drinking glass, two urine jars, and a vessel 
for defecation. The balcony floor is 2.5 meters from the calorimeter 
room floor; the stairs leading to it have a rather sharp inclination, with 
12 steps. (See Plate 3, figure E, page 31.) Before arranging to have 
him occupy this balcony, he was asked if he would be likely to become 
dizzy as a result of going up and down the stairs, but he replied that he 
was never dizzy during his fasts. Indeed, he seemed to like the idea 
of living in this balcony, as it gave him considerable freedom and yet 
made it possible to watch him. Plate 1, figure A, gives a view of him 
in a characteristic pose, writing at his desk in the balcony. 

Throughout the fast he was under constant surveillance by various 
responsible members of the staff and there were nearly always two or 
three assistants on duty in the room. It was therefore impossible for 
him to leave the balcony or to obtain food without its being known 
at once. The watching problem in this experiment was very simple. As 
L. was unknown in this country, he had no friends who would attempt 
to bring food to him and all of those who came into communication 
with him had a scientific interest in having the fast carried out to the 
end of the time planned. While he might have drunk his urine, as did 
the faster Jacques, it would have been practically impossible for him 
to do this without its being known. Moreover, he had too much inter- 
est in the fast to do anything of the kind, and we firmly believe that if 
he had been surreptitiously offered food, he would have refused it. 

29 



30 A STUDY OF PROLONGED FASTING. 

The three days preceding the fast — the so-called preliminary period — 
were used to accustom the subject and the staff of assistants to the 
apparatus and to the general routine, in order that the program could 
be carried out as smoothly as possible, and without too great a demand 
upon the time of the physicians and co-workers who made observations 
upon the subject. His diet and daily life were under constant obser- 
vation during this period, but he was free to choose his food and to 
arrange his time as he desired when no tests were being made upon him. 

It was necessary to be certain that L. was physically and psychically 
a fit subject for the long fasting experiment. He was accordingly given 
several rigid physical examinations by Dr. H.W. Goodall, of the Harvard 
Medical School, and also underwent a psychical examination by Prof. 
E. E. Southard, director of the Massachusetts Psychopathic Hospital. 
The results of these examinations gave us every assurance that L. 
was a suitable subject for this long fasting experiment. 

The body- weight of this subject when he reached Boston was some- 
what smaller than the initial body- weight reported for his earlier fast. 
L. stated that his body-weight at the beginning of the previous fast 
was excessive and that he desired to begin this experiment with his 
normal body-weight. While this reasoning was scientifically correct, 
his small weight caused us considerable anxiety, as it was feared that 
he would be unable to endure a 31-day fast. Inasmuch as a fast of 
7 to 10 days' duration would be of practically no value to us except as 
a duplication of the earlier work, every possible arrangement was made 
to adjust the conditions so as to prolong the fast; the subject quickly 
found that if he made the statement that any particularly distasteful 
routine or test would tend to "shorten the fast," it would be omitted. 
On the other hand he took an intense interest in the outcome of the 
experiment and had an almost religious belief in the benefits to 
humanity to be derived from it. He enjoyed the distinction of having 
so many observers studying him, and his peculiar appreciation of the 
scientific value of the observations enabled us frequently to induce him 
to waive his objections to any routine by a summary refusal to go on 
with that particular test unless the routine were carried out. To his 
credit it must be said that whatever idiosyncracies he exhibited at 
times, he would, after reflecting on the importance of the experiment, 
beg for the continuation of the complete routine. 

To secure as much information as possible regarding the normal 
metabolism of L., he was asked to sleep inside the respiration calo- 
rimeter immediately on arriving in Boston. He was provided with a 
comfortable bed, air mattress, and bed clothing, the bed comparing 
well in size and comfort with a berth on an ocean steamer. Important 
data regarding the normal metabolism of this subject were thus secured 
for several nights before the actual fast began. Fortunately L. slept 
very quietly, and when not asleep he remained very quietly in the 
same position for long periods of time, thus greatly facilitating the 
accurate measurement of the metabolism. 



PLATE 3 




E. L. on the Thirty-first day of the Fast, ascending the Stairs of the Balcony. The picture required 
20 seconds ; there is no Evidence of Unsteadiness. 




F. Clinical Examination by Dr. H. W. Goodall. This photograph was taken on the Thirty- first day 
of Fasting, upon the Balcony occupied by L. during the day. 



GENERAL HISTORY OF FASTING EXPERIMENT. 31 

PROGRAM FOR RESEARCH. 

The many observations and the large number of co-workers and 
assistants made a carefully prepared program absolutely essential, so 
as to use the actual available time of the subject and the co-workers to 
the best advantage. The observations planned for each day were the 
weighing of the subject after he had urinated and arisen; blood tests; 
measurements of blood-pressure and the alveolar air; test for acetone 
in the breath; records of rectal temperature and of pulse-rate; the 
careful collection, measurement, and subsequent complete analysis 
of the urine; and the apportionment and measurement of the water 
taken. The subject entered the bed calorimeter about 8 o'clock each 
night, remaining there until 8 o'clock the next morning, during which 
time the respiratory exchange, water vaporized, and heat produced 
were continuously measured. He was then taken out and his respira- 
tory exchange was observed in three experimental periods by means 
of the universal respiration apparatus. (See Plate 2, fig. C, page 19.) 
Respiration experiments were also frequently made with the subject 
at other times of the day and in varying body positions. The respira- 
tory exchange when the subject was breathing an oxygen-rich atmos- 
phere was determined several times, and a series of respiration experi- 
ments was made by Mr. T. M. Carpenter when the subject was writing. 
(See Plate 1, figure B, page 11.) In addition to the regular routine, 
there was a rigid clinical examination by Dr. Goodall every second 
day (see Plate 3, figure F), psychological tests were made by Dr. 
H. S. Langfeld, and anthropometric measurements were taken by 
Professor W. G. Anderson once a week. Every five or six days a 
complete series of photographs was made of the naked subject. (See 
Plates 4 and 5, p. 65.) Once a week his body was washed with distilled 
water, the water used being preserved and analyzed. Among the many 
incidental observations carried out during the experiment was a series 
of X-ray plates on the thirtieth day of the fast by Dr. F. H. Williams 
and a study of the flora in the colon on the thirty-first day by Dr. A. I. 
Kendall. In clear, pleasant weather the subject was taken to the 
roof or more frequently given a drive through the park system of Boston. 
The program for a typical day — that of May 7-8, 1912 — appears below: 

May 7. 7 h 46 m a.m. Bed calorimeter experiment ended. 

Respiration experiment (three periods.) 

Weighed. 

Photographs taken. 

Blood sample taken. 

Blood pressure tests. Alveolar air. 

Respiration experiment made with subject 
writing (two periods.) 

Psychological tests. 

Respiration experiment (two periods). 

Bath of distilled water; underwear changed. 

Entered bed calorimeter. 

Calorimeter experiment begun. 
May 8. 7 50 a.m. Bed calorimeter experiment ended (5 con- 

secutive periods.) 



7 h 46 ir 


1 a.m. 




8 15 


a.m. 


to 9 h 17 m a.m. 


9 28 


a.m. 




10 00 


a.m. 




10 30 


a.m. 




1 40 


p.m. 




3 43 


p.m. 


to 4 h 14 m p.m. 


5 00 


p.m. 




7 01 


p.m. 


to 7 h 44 m p.m. 


7 50 


p.m. 




8 23 


p.m. 




9 34 


p.m. 




7 50 


a.m. 





32 A STUDY OF PROLONGED FASTING. 

It will be noted that the regular clinical examination did not take 
place on this day, as these examinations were made only on alternate 
days. Furthermore, no drive was taken. As will be seen from this 
typical program, the subject found himself fully occupied by the vari- 
ous observations; in all of these he took a keen personal interest. 

DAILY RECORDS OF FASTING EXPERIMENT. 

Although a definitely arranged program was prepared, and for the 
most part rigidly followed, the daily routine was varied by a large 
number of extraneous observations, particularly in regard to the feel- 
ings and moods of the subject, as well as observations made by co- 
workers. To present these adequately, it seems desirable to give 
them in the form of a daily record beginning with the arrival of the 
subject at the Nutrition Laboratory. This record will be in the nature 
of a "log-book," which will simply give the general history of the 
experiment from day to day, with no attempt to describe the technique 
or discuss the results. 

The experimental day for most purposes ended with the completion 
of the respiration experiment at about 9 h 30 m a. m. The last meal 
was eaten at 6 p. m. on April 13, 1912; thus the true fasting period 
began at 9 h 30 m a. m., April 14, or about 15 hours after the last meal. 
In this daily history the personal observations of the subject on his 
experiences during the night are always given in the notes for the next, 
day, but all events up to the moment of entering the calorimeter are 
recorded on the date of occurrence. While much that is said regarding 
the previous fasts of the subject must, from a strictly scientific stand- 
point, be considered as worthless, yet the trend of thought is not 
without interest in interpreting the mental make-up of the subject. 

PRELIMINARY PERIOD. 

April 10, 1912. — L. arrived at the Nutrition Laboratory about 8 p. m., 
coming directly from the steamer and leaving his baggage on the wharf, but 
bringing with him the samples of urine which he had collected for several days 
on his passage across the ocean. He showed himself to be heartily in sympathy 
with the plan of the experiment, and appeared to be a subject who would 
cooperate fully in the experimental routine; he placed himself entirely in our 
hands. As he had had no evening meal, he was sent to a hotel with one of the 
laboratory assistants. This meal, which was of his own selection, consisted 
of a large beefsteak with onions, one boiled potato, one portion chocolate ice 
cream, and a glass of water. 

He returned to the laboratory at 10 o'clock, and then reported that he had 
had a very rough passage on the steamer, there being but a few hours of smooth 
sailing on the third day out and a few hours on the last day. On the other 
hand, he appeared to be in very good condition and showed no bad effects from 
the discomforts of the trip. He maintained that during the previous year he 
had lived almost exclusively on a vegetarian diet, taking occasionally milk and 
cheese, very rarely eggs, and no meat. On the steamer, however, the menu 
did not include the food he was accustomed to and he was compelled to eat 



GENERAL HISTORY OF FASTING EXPERIMENT. 33 

meat and "highly seasoned sauces" which he did not particularly care for. 
During the previous year he had been living upon one meal a day, which was 
eaten about noon, using the juice of an orange to satisfy his thirst when 
necessary. He claimed that this limited dietary had been very beneficial to 
his health. During the fast he wished to drink distilled water. He did not 
especially like it, and usually drank hot water, but since some people believed 
that there was nutriment in water, he wished to use distilled water so that there 
could be no question as to his obtaining nutriment in this way. 

In his previous fasts it had been his custom to carry on his regular business 
and to go into court and plead his cases as usual, thus engaging in a not incon- 
siderable amount of muscular activity. He snowed a decidedly intelligent 
interest in the experiment, as was indicated by his asking if the eyes should not 
be examined by an eye specialist, for he had found that as a fast progressed, 
the eyesight improved considerably, though normally he had very poor eye- 
sight. He also thought it important to study the blood and seemed much 
gratified when he was told that both eye and blood tests would be made. The 
important role which he would play in the experiment was emphasized to him 
and he was shown that the efforts of the laboratory staff would be of no avail 
without his full cooperation. His attitude toward the experiment and under- 
standing of the requirements showed him to be by far the most intelligent 
man that has ever been studied as a fasting subject. 

When discussing the question of defecation during a fast, he made the state- 
ment that in some of his long fasts he had defecated only once or twice. Often 
he defecated shortly after the beginning of the fast and then not again until 
after the fast was over, but after beginning eating he defecated quite regularly. 
In one fast he said that he did not defecate until the twenty-seventh day. 

As there was no time that evening to discuss with him at length his past 
hiBtory and the details of the fasting experiment, he was taken down to the 
calorimeter laboratory, where he urinated, removed all but his underclothing, 
and prepared to go into the calorimeter. The stethoscope was adjusted, and 
the rectal temperature taken with a clinical thermometer, which was left in the 
rectum three minutes. He drank a glass of water and was then placed inside 
the chamber of the calorimeter. After he had been shown how to use the tele- 
phone and the signal bell, a black cloth was placed before the window so that 
the electric light would not disturb him and the calorimeter was then sealed. 

Even on this first day the subject was inclined to talk about the method of 
breaking his fast, saying that he was accustomed to do this by taking the juice 
of one or two lemons, and afterwards orange juice, to which he sometimes 
added sugar. As was seen later, such a method for breaking the fast proved 
to be disastrous to our predetermined plan of securing data after the fast. 

April 11, 1912. — The calorimeter experiment for the previous night was 
uneventful and ended at 8 h 02 m a.m. The measurements were made in three 
consecutive periods, the idea being to secure observations during the latter 
part of the night and thus eliminate, if possible, the influence of food taken 
during the evening. The subject kept very quiet most of the time, and proved 
exceedingly tractable and intelligent. The importance of lying quietly inside 
the respiration chamber had been impressed upon him and we have rarely had 
a subject who lay so quietly for so long a time. About 5 h 30 m a. m. he tele- 
phoned to ask if everything were all right, reporting that he felt very well but 
had been awake for some hours. He was instructed to ring an electric bell 
every few minutes by pressing a small push-button inside the chamber, so 
as to show us that he was awake. He rang this bell regularly throughout 
the rest of the experiment, beginning at 5 h 30 m a. m. 

When the calorimeter was opened, a strong odor of onions was apparent, 



34 A STUDY OF PROLONGED FASTING. 

doubtless due to the fact that he had eaten beefsteak and onions the evening 
before. L. reported that for the first two hours after he entered the chamber 
he was very warm, but later became cool and slept comfortably. Since the 
temperature inside the calorimeter seldom varies by 0.1° C, such an observa- 
tion serves excellently to illustrate the futility of placing any weight on personal 
impressions. The interior of the respiration chamber reminded him of his 
cabin on the steamer. It was impossible to obtain records of the pulse-rate 
for about one hour after 5 h 30 m a. m., this being due to some change in the 
position of the stethoscope. Later the subject was able to readjust it and the 
records were obtained thereafter. 

When L. came out of the chamber, shortly after 8 a. m., he urinated and 
immediately the experiment with the universal respiration apparatus was 
begun. (See Plate 2, figure C, page 19.) This experiment consisted of three 
15-minute periods. 

Almost the entire day was spent by the subject in familiarizing himself with 
the experimental routines. After the respiration experiment was over, several 
tests of the blood pressure were made, and samples of the alveolar air taken by 
the Plesch and Haldane methods. Professor W. G. Anderson made a series 
of physical measurements and attempted the routine strength tests, but 
was not able to obtain these, owing to the disinclination of the subject. It did 
not seem advisable to complicate the program by taking photographs on this 
day. An examination of the blood was made, also a most careful clinical 
examination. L. then took a hot bath and went out with Mr. H. L. Higgins 
for the first meal of the day. This meal, selected a la carte, consisted of one 
portion of scallops and tartare sauce, one portion of roast lamb and mint sauce, 
two portions of mashed potato, three rolls, two portions of butter, and one 
portion of custard pie. 

On returning to the laboratory in the afternoon, he occupied himself in 
writing letters and in talking with different members of the staff until about 
4 h 30 m p. m., when the first series of psychological tests was made. The visual 
acuity test was not very successful, as L. has a very short vision and the letters 
used were so small that new ones had to be secured. The chief value of the 
test on this day was to familiarize the subject with the routine. L. continued 
to have a keen interest in the success of the experiment and cooperated in 
every way except in the strength test. 

In order to make sure that the last meal of the day should contain only a 
small amount of protein, I went personally with him to the restaurant. His 
supper at this time consisted of half a grapefruit, to which he added quite a 
little sugar, a plate of split-pea soup (this containing practically all of the 
protein in the whole meal), two or three slices of bread and butter, one portion 
of stuffed tomatoes, one of fried sweet potatoes, another of white potatoes, 
one dish of strawberries and cream, and some strawberry ice cream. He 
returned to the laboratory about 9 h 30 m p. m. and prepared for the night in 
the calorimeter. 

While I was with him in the afternoon and evening, L. gave me considerable 
information regarding his previous fasts. Although the unscientific nature of 
these personal impressions is recognized, it seems desirable that they should 
be recorded. His observations for the most part had to do with defecation, 
feelings of hunger, and changes in body-weight. During his fast of 40 days, 
which he reports as having been broken on April 10, 1911, he defecated twice 
during the first two days, and again, according to his remembrance, on the 
twenty-fifth day. The feces on the latter day consisted of a small amount of 
blackish or very dark brown material, with a yellowish white mucus. He 
defecated again the evening of the day on which he broke the fast. He 



GENERAL HISTORY OF FASTING EXPERIMENT. 35 

noticed that there was a large amount of gas in the intestines. His theory was 
that the lower portion of the bowels was clogged with feces, the rest of the 
intestines being filled with air, and that when he ate, the air was compressed 
by the food passing along the intestines, this compressed air distending the 
bowels and producing much colic. Thinking that it might be more advan- 
tageous for him to empty the lower bowels by an enema before beginning the 
fast, I suggested this to him, but he preferred not to take an enema unless 
it were scientifically necessary, as he believed in natural rather than forced 
movements. He had never had any distress from defecation or from inability 
to defecate such as that experienced by the subject of the fasting experiments 
made in Middletown, Connecticut. He also said that he usually had no hunger 
pains, but felt somewhat hungry. His weak point was his throat, which fre- 
quently troubled him considerably. On the twentieth day of his long fast 
his throat was very dry and slight traces of blood appeared. On the second 
day of the same fast he noticed that he was a little irritable. In the first 
part of a fast he was usually somewhat depressed, but not to any great degree, 
and after the third day he would be very happy, with no desire for food. 
During one of his Malta fasts he went to the table every day and watched 
his children eat, and, in fact, prepared delicacies for them. The first indication 
of a desire to break the fast was usually shown by an intense craving for an 
acid and it was his custom to take first a lemon and then an orange. All 
throughout his earlier Malta fast the color of his tongue was unnatural. 

The changes in body-weight during his previous fasts were most significant. 
The initial weight of the 40-day Malta fast was much greater than when he 
came to Boston and even at the end of the fast he weighed more than he did 
on this date. His contention was that when a man fasts with a large amount of 
fatty tissue, he is "really not fasting but simply draws upon body tissue, so 
that the only true fasting is when a man begins the fast with a normal weight." 
If I had had the decision of the matter, I should have preferred to have him 
begin his fast with a larger amount of fatty tissue than he had. Nevertheless 
the results have more interest from the fact that he did not have this excessive 
amount of fat. 

He continued to keep up his interest in the results and wished to know them 
from day to day, but I pointed out to him that as the fast progressed, if the 
results were abnormal in any way and he should be told of them, he might 
instinctively and unintentionally attempt to alter the conditions so as to meet 
the variations that we should find. It was suggested to him that he should 
spend all of his energy and interest on familiarizing himself with the technique 
and pay little attention to the results that we found, until the fast was over. 

On this day he discussed extensively the writers on fasting, more especially 
those which would be designated as the semipopular writers. He pointed out 
rathei naively that most of the writers on fasting wrote of the experiences of 
others, but never fasted themselves. In speaking of the fasts carried out by his 
wife, he said that during a prolonged fast menstruation disappeared entirely. 

L. was quite impressed with the increased mental activity and power of 
working during a fast and maintained that any student who is to take an 
examination should know better than to take an excessive amount of food. 
He cited instances where he wrote poetry and continued his literary work 
during his fast and said that he was conscious of a considerably increased 
efficiency. According to his experience, the hours of sleep decreased somewhat, 
and although he began work every morning during the fast at 5 o'clock, he 
felt but little fatigue. 

He gave no special attention to the amount of muscular exercise taken, but 
walked to and from his business, pleaded cases in court, went to his club, and 
walked about the street as usual, this exercise continuing some six or eight 



36 A STUDY OF PROLONGED FASTING. 

hours each day, but he took no very long walks. He said that at one time, 
when he was being jeered at in regard to his fasting by one of his clubmates 
who said that he was losing strength, he suggested that they test his strength 
with a hand dynamometer, and he was able to show more pressure on the hand 
dynamometer than the man who ridiculed him. He did not believe in Succi's 
theory, however, that there is an increase in muscular strength during a fast. 

April 12, 1912. — The records of the night observers showed that L. slept 
rather quietly from 9 h 40 m p. m. on April 11 until 3 h 33 m a. m. on April 12, 
after which time he rang the bell at intervals from 5 to 10 minutes. At 5 >, 30 m 
a. m. he telephoned that it was necessary to urinate. For this he used a 500 
c.c. urine jar which had been placed inside the calorimeter the night before, 
but notwithstanding its size it was not sufficiently large to contain all of the 
urine he desired to pass. The subject reported no particular discomfort from 
the fact that he could not completely empty the bladder. 

At noon on this day, L. went to dinner with Mr. T. M. Carpenter. His 
dinner, selected by himself, consisted of one portion of broiled salmon and a 
small quantity of green peas, two pork chops saut£, three heaping tablespoon- 
fuls of mashed potato, a dish of sliced cucumbers and tomatoes, three small 
rolls, two small pieces of butter, one portion of strawberry ice cream, and two 
glasses of water. 

At 7 p. m., he took supper, likewise with Mr. Carpenter, eating one large 
portion of macaroni, with apparently but little cheese in it, one large portion of 
fried sweet potatoes, and one large portion of fried eggplant. In addition, he 
ate two slices of French bread, two portions of butter, one portion of chocolate 
ice cream, one-half dozen macaroons, and drank five glasses of water. He 
seemed to enjoy this meal very much. 

Foi the first time on this day he showed apprehension in regard to the 
experiment, maintaining that the number of tests made with him would tend 
to shorten the fast, as it required concentration on his part to cooperate with 
the different observers and this concentration used a certain amount of his 
energy. On the other hand, he suggested that a series of anthropometric 
measurements should be added to the tests planned for him in order to show 
that he was a normal individual, citing the fact that one objection that is 
made to professional fasters and men who fasted any great length of time is 
that they are not normal people intellectually and for that reason the results 
obtained with them could not be considered normal. He considered these 
measurements of the greatest importance and thought they could be obtained 
by taking a photograph showing the angle of the face, width of the head, etc. 
Since he desired them, a series of these photographs was subsequently taken. 

April 13, 1912. — The records for this day show that the subject entered the 
calorimeter chamber at 9 h 40 m p. m., April 12, and apparently was quiet until 
12 h 17 m a. m., April 13, when he began ringing the bell and continued this more 
or less regularly, i. e., several times an hour, throughout the remainder of the 
night. He telephoned at 4 h ll m a. m., and again at 5 h 10 m a. m., stating 
that he had slept little but felt very comfortable. He urinated at 4 h ll m a.m., 
passing 595 c.c. of urine. On coming out of the calorimeter, he had consider- 
able to say regarding the dreams that he had had; between 9 h 40 m p. m. and 
12 h 17 m a. m., he slept very well without dreaming. After that period his 
sleep was much broken, and he had a number of dreams, one of which was 
accompanied by a seminal emission. He explained this by saying that his 
supper the night before contained too large a proportion of carbohydrates, 
which heated his blood and made him very uncomfortable. He almost imme- 
diately remarked that he had been very comfortable inside the chamber and, 
in fact, rather enjoyed being there. 



GENERAL HISTORY OF FASTING EXPERIMENT. 37 

A considerable number of photographs were taken of the nude subject, a 
procedure which evidently gave him much pleasure (see Plates 4 and 5). 
At ll h 16 m a. m., he urinated and defecated, and later went out to his midday 
meal with Mr. Higgins. He selected pea soup with three crackers, one gen- 
erous portion of fish (finnan haddie), mashed potatoes, two lamb chops, one 
portion of French fried potatoes, four slices of bread with butter, and one 
portion of strawberry ice cream and fruit syrup (college ice). 

In the forenoon of the third day I told the subject that it was important 
that we should have some idea as to how long he expected to fast, for he had 
frequently made the statement that so many tests upon him would tend to 
shorten the fast. He was shown that a short fast would have no interest for 
us. He replied that he expected to fast until his body-weight fell below 100 
pounds (45.4 kilograms). Inasmuch as his initial weight was 134 pounds 
(60.6 kilograms), he thought he would lose about a pound a day, which would 
make the fast approximately 30 days long. 

He brought me two sealed bottles of the liquor that Succi uses in his fasts to 
allay the pangs of hunger at the beginning of a fast. One of these bottles he 
presented to the Laboratory, and asked me to keep the other for him until the 
fast was concluded, as he did not wish any one to say that he took the liquor 
and that it helped him to carry out the fast. The subject sympathized fully 
with the strictest surveillance and was much impressed with the fact that so 
many co-workers were watching and studying him. 

In order to obtain an expert opinion regarding the mental state of the subject 
at the beginning of the fast, arrangements were made by which Dr. E. E. 
Southard, of the Massachusetts Psychopathic Hospital, could examine him 
at 3 h 30 m p. m. on this day. 

L. was excessiuely voluble, continually emphasizing the importance of making 
measurements of his face and head, the length of the ears, and similar measure- 
ments, as he desired that a careful study should be made in order to prove 
that he was a normal man, and not erratic and abnormal. He also advocated 
with great persistency the study of the influence of fasting on the sexual organs. 

At 6 p. m. he went with Mr. Carpenter to a local restaurant for the last meal 
before beginning the fast. This meal consisted of bananas and cream, straw^ 
berry shortcake, ice cream, and three glasses of water. 

In the evening he took a hot bath, after which he was sponged with distilled 
water, and then put on a union suit and a pair of white stockings, both of which 
had been previously thoroughly washed and rinsed in distilled water and dried. 
The union suit absorbed the perspiration, so that a measure could be obtained 
of the nitrogen eliminated through the skin in the form of urea or organic 
nitrogenous material. After drinking a glass of water, he entered the bed 
calorimeter at 9 h 44 m p. m. 

FASTING PERIOD. 

April 14,1912 (first day of fast). — According to the experimental records, 
L. rang the bell to show that he was awake from time to time during the night, 
there being one hour of quiet between 10 h 17 m p. m. and ll h 29 m p. m. and 
another hour between ll h 48 tt p. m. and 12 h 49 m a. m., but from 2 h 06 m a. m., 
April 14, he rang the bell more or less regularly until 5 h 25 m a. m., when for 
two hours he remained quiet and was apparently asleep. At 2 h 06 m a. m. he 
telephoned that he was compelled to urinate. This disturbed him, as he had 
been sleeping very soundly. Later he suggested that if he could drink water 
during the day and not have to urinate during the night, he would feel more 
comfortable. His program was therefore arranged subsequently, to include 
the taking of the greater part of the water during the daytime. 



38 A STUDY OF PROLONGED FASTING. 

When he came out of the calorimeter, he reported that he felt fairly com- 
fortable. Some of the air had been let out of his air mattress, so that he found 
it easier than formerly. After the respiration experiment was over and he had 
washed his hands and face, he went up into the balcony and a thorough 
examination was made of all his clothing and baggage. This was done to 
make sure that nothing was hidden, such as food tablets, or alkaloids of any 
kind, in a form that might be sewed into his clothing, concealed in hollow 
books, in the tips of his shoes, or in some similar place. As he had had previous 
experience as a pharmacist, it seemed desirable that such precautions should 
be taken. Even the linings of his clothing were examined, but nothing was 
found, only two small cakes of soap being removed from the balcony. Besides 
his clothing and numerous testimonials from many of his Malta associates, a 
large part of his luggage consisted of so-called "fasting literature"; of particu- 
lar interest to us were the materials he had collected in his visit to Succi on his 
way to Boston. 

He had never been in the habit of using his eyes for reading or studying in 
the evening, but finished his work by 6 or 7 o'clock and rose early in the morning. 
It had been his custom to spend the evening with his family or go out for a 
walk, or to some place of amusement. Anticipating the tediousness of the 
fast, we sought to interest him in some simple game as solitaire, checkers, or 
whist, but he refused all of these and preferred to retire early, entering the 
respiration calorimeter even earlier than we had planned. 

Much of the first day he talked about how well he would feel as a result of 
his fast, how happy he was to think that he was beginning it, how he would be 
relieved from the necessity of eating and drinking, so that the time he now 
spent in this way could be devoted to higher mental work. As he usually felt 
so much better during a fast, he expected that these fasting days would be 
among the happiest of his life. He considered that he bore the tests on this 
day much better than previously, showing a greater power of concentration. 
In the psychological test he observed that the ticking of the metronome 
seemed louder than in previous tests and attributed this to the fact that during 
fasting his hearing was always more acute. Inasmuch as this observation was 
made when he had omitted but one meal — the meal in the middle of the day — 
the weight which should be given to this observation is easily estimated. 
Although the weather was dull, which usually depressed him, he was perfectly 
certain that throughout this day he felt much better both mentally and physi- 
cally than the day before. During the evening he was unusually lively and 
cheerful, sang and whistled quite a little, and said he felt like dancing. 

Although the temperature on the balcony was 22° C, he began to complain of 
the cold and in the afternoon wore a blanket wrapper and bedroom slippers in 
addition to his regular clothing. (See Plate 1 , figure A, page 11.) Subsequently 
he wore his heavy-weight suit of underclothes over the union suit and stockings, 
which had been washed in distilled water. 

He said that he never used alcohol in any form and in the afternoon com- 
plained of having to drink too much water, remarking that it would shorten 
the fast if he drank so much, as it would wash out the salts. While there was 
a legitimate foundation for the statement that the distilled water might affect 
the salt metabolism, it was evident that L. had discovered an efficacious way of 
obtaining anything that he wanted by bringing forward the argument that it 
would "shorten the fast." It was decided that if the amount of water given 
him to drink caused him discomfort, he could lessen the amount. He pre- 
ferred to drink only when thirsty, but that morning had taken water without 
feeling the need for it, and had had absolutely no hunger all the day. The 
water tasted much better than he had expected it to taste, as he usually dis- 
liked the taste of distilled water. 



GENERAL HISTORY OF FASTING EXPERIMENT. 39 

In preparing him for the respiration chamber at night, it was difficult to 
adjust the stethoscope so as to hear the pulse-beats through it clearly. The 
best results were obtained by placing the stethoscope about 1 cm. above the 
left nipple and 2\ cm. toward the center line of the chest. He urinated at 
6 h 20 m p. m. and thought he would not urinate again during the night. By 
this time he had become thoroughly accustomed to the calorimeter, showing 
no anxiety regarding his stay in it and having the greatest confidence in those 
who had charge of the experiment. He entered the respiration chamber at 
8 h 48 m p. m. 

April 15, 1912 {second day of fast). — From the experimental records and the 
report of the subject, L. evidently had a very comfortable night, ringing the 
bell only occasionally and sleeping much better than he had any night since 
he had come to Boston. When he left the apparatus, he reported himself to 
be in excellent condition. There was no noticeable odor when the calorimeter 
was opened. He felt no pain or sensation of hunger, and very little thirst. 
During the forenoon he wrote busily and several long letters were mailed for 
him at noon. L. is ambidextrous, using his right hand for writing and left 
hand for work, that is, in taking the dynamometer test, he always used the left 
hand first. In conversation, he invariably led up to the discussion of the 
innumerable "popular" books on fasting, with which he was remarkably 
familiar. An ingenious argument against fasting, which he reported as having 
been given him by a prominent American vegetarian, was that in a fast a man 
became a flesh-eater, as he existed upon his own flesh. 

At 4 p. m., 250 c.c. of water were taken from the 1 liter of distilled water 
contained in his bottle, reducing his apportionment of water to 750 c.c. 
Subsequently this amount was given to him daily. The rectal thermometer 
was used for the first time during the night of the second day of fasting, L. 
inserting the thermometer himself. Previous to entering the calorimeter, the 
subject was in unusually good spirits, singing and talking a great deal. 

April 16, 1912 {third day of fast). — On coming out of the apparatus in the 
morning, L. said that he wore the rectal thermometer all night and suffered no 
distress, although at times it troubled him somewhat. The observer reported 
a good series of temperature measurements throughout the night. L. evidently 
slept more soundly than usual, ringing the bell only occasionally. No special 
odor was noted when the chamber was opened. 

After the subject had been weighed, several photographs were taken of him 
nude in various positions, corresponding approximately to those taken on 
April 13. After he was dressed, at his request several photographs were taken 
of the head to show the facial characteristics. His innumerable suggestions 
displayed a worthy interest in the experiment, though many of them had to 
be disregarded. 

On this day the drinking water and the urine bottle were placed on a table 
at the foot of the stairs leading to the balcony, the subject notifying the 
assistant whenever he wished water or the urine bottle. In this way all 
possibility of his drinking the urine, as was done by the subject of Paton and 
Stockman, was eliminated. Although L. was less cheerful than he was the 
day before, he was by no means depressed, spending a considerable part of the 
forenoon in writing in his diary. He was particular ly cautioned against writing 
any of his experiences or sending out information unauthorized, as misuse 
would be made of the material which he gave out before the fast was completed. 

We had expected that he would lie down occasionally and had provided a 
well-upholstered couch for his use, but he sat up practically all day. It seemed 
desirable, therefore, to study his metabolism in this position. Accordingly in 
the afternoon he came down from the balcony and a respiration experiment was 



40 A STUDY OF PROLONGED FASTING. 

made with him in two experimental periods, while he was sitting in a com- 
fortable chair. 

During the day the subject reported that he had no bad feelings of any kind. 
He felt slightly irritable, but considering the confinement, he was remarkably 
free from this feeling of irritation, much more so than in his last fast. The 
disinclination to exert himself in any way which might cause the slightest 
strain was shown on this day when he complained that the dynamometer hurt 
his right hand and he did not dare to press it as hard as he would have liked 
to. He passed no urine from 8 h 05 m a. m., when he came out of the bed 
calorimeter, until 8 h 05 m p. m., just before entering the chamber, although he 
drank all of the 750 c.c. of distilled water, taking the last portion at 5 h 50 m p. m., 
when he finished the psychological tests. 

In the evening, just before entering the calorimeter, he said that the rectal 
thermometer irritated him considerably and had kept him awake more or less 
the night before. He was aware of its presence every time he woke up and 
during the day the anus had been somewhat irritated. He asked if the temp- 
erature could not be taken every other night instead of every night, as had been 
planned, and the thermometer was accordingly not used on the night of 
April 16-17. 

April 17, 1912 {fourth day of fast). — Both the experimental records and L.'s 
report indicate that he slept better on the night preceding this day than he had 
any night thus far passed in the chamber. The day was uneventful. He said 
that he had no great thirst and he drank the distilled water more from a sense 
of duty than from any desire for it. He did not leave the balcony during the 
day except for the psychological test at 4 h 50 m p. m. At one time during the 
forenoon he was seen to hold up a light man, weighing but 125 pounds (56.7 
kilograms) for a few seconds. He appeared pleased with this supposedly 
"sensational feat" and was quite unappreciative of the caution to conserve 
his strength for tests of muscular strength that could be measured, tests that 
he had repeatedly refused to make. 

On this day he wore the stethoscope most of the day, an assistant counting the 
pulse-rate at a distance from him and unknown to him for the greater part of 
the time. It was hoped that more or less continuous records of the pulse-rate 
could be obtained in this way. In preparing him for the calorimeter at night, 
I personally inserted the rectal thermometer and he reported that it did not 
hurt him in any way. He believed that, when putting it in himself the first 
night, he must have irritated the anus somewhat, thus causing the subsequent 
discomfort. The subject entered the calorimeter for the night at 8 h 19 m p. m. 

April 18, 1912 {fifth day of fast). — The subject reported in the morning that 
he slept quite well during the night, but not so well as he did the night before. 
He thought he went to sleep about an hour after entering the calorimeter and 
woke up several times during the night. He said he was entirely comfortable, 
the rectal thermometer giving him no trouble. The spring supporting the 
bed had been somewhat weakened in order to obtain a greater sensitivity for 
the apparatus recording the muscular activity. The records show, for the 
most part, a remarkably quiet night, so that long calorimeter experiments of 
10 or 11 hours will be perfectly comparable so far as the muscular activity is 
concerned. In this respect the subject was exceptionally well adapted for an 
experiment of this kind. 

Notwithstanding the dull weather, L. said that he felt very well, with no 
loss of strength, headache, feeling of hunger, or of apprehension. He was not 
usually troubled with headache during a fast. When his measurements were 
taken, his height was found to be 1.707 meters. He said that his maximum 
weight was 14 stone or 196 pounds (88.9 kilograms). His initial weight for 



GENERAL HISTORY OF FASTING EXPERIMENT. 41 

the Malta fast was 12 stone and 3 pounds, or 171 pounds (77.6 kilograms). 
During this fast he lost 37 pounds (16.8 kilograms), and thus vveighed at the 
end of the fast 134 pounds (60.8 kilograms). These measurements included 
the weight of the clothing worn and were taken with ordinary scales, but 
probably represent average weights. 

In the evening L. talked about his work in Malta and his interest in the 
numerous fasting books which he had read. He became quite excited in 
talking of three legal cases in which he was involved in Malta, speaking with a 
good deal of vigor and enthusiasm. There was not the slightest evidence of 
his having fasted for 4§ days. After inserting the rectal thermometer, the 
subject entered the bed calorimeter about the usual time. 

April 19, 1912 (sixth day of fast). — The subject was sleeping very soundly 
when the calorimeter experiment was ended and on leaving the apparatus, 
reported that he had had a very comfortable night. He seemed very bright 
and quite like himself. According to the records of the observers, the subject 
slept very well throughout the night until 5 h 30 m a. m., when he rang the bell 
intermittently for a short time. He still continued to have no discomfort, no 
sensations of hunger, and no apprehensions, and felt very much inclined to 
mental work. He said he did not expect to have such an excellent fast and 
seemed to be content with the progress of the experiment. On this day the 
subject spent considerable time in writing. To provide data for subsequent 
computation of the total energy transformation during the 24 hours, it seemed 
desirable to measure the respiratory exchange while the subject was writing. 
This type of experiment, which had never been attempted in this laboratory 
before, proved to be very successful, the subject sitting in a comfortable chair 
and writing as usual. (See Plate 1, figure B, page 11.) 

April 20, 1912 (seventh day of fast). — When the calorimeter was opened, no 
odor of acetone was apparent, and no unpleasant odor of any kind. Thus far 
in the series of calorimeter experiments, the only unpleasant odor which has 
been noted was the strong smell of onions on the morning following his evening 
meal of beefsteak and onions. L. said that he was in excellent condition; 
intellectually he was very acute. He slept well throughout the night with no 
discomfort and rang the bell only a few times. With the sunny weather the 
subject's spirits rose and he reported himself as feeling very happy and 
uplifted; he sang at times during the morning. A barber cut his hair and 
beard on this day, the trimmings being saved for analysis. 

After taking a bath in the evening, he entered the calorimeter. He told 
one of the assistants that he felt very drowsy and expected to go to sleep 
almost immediately. 

April 21, 1912 (eighth day of fast). — On coming out of the apparatus, the 
subject reported that he did not go to sleep as quickly as he had expected to; 
he lay awake for 2 hours and also woke up in the night once or twice. He 
slept well, however, and had several dreams which he reported to Dr. Langfeld. 
L. asked if he did not have fever during the first 2 or 3 days of the fast, stating 
that his wife during the first days of her fast had quite a high fever. When 
he was told that he had had no fever he was much surprised. His theory was 
that "in the first days of the fast, the blood rid itself of all impurities, the 
burning process producing a high fever." 

In the afternoon, the subject went up to the roof, using the elevator; he 
remained there for about 1? hours, sitting in the sun and chatting on general 
subjects. He said considerable about his experience with Esperanto and about 
many of his Maltese customs. His voice was strong, and he seemed to be 
very well and bright. He came down from the roof by the elevator to the 
office and dictated into the dictaphone a statement regarding his feelings. 
The dictated report is as follows: 



42 A STUDY OF PROLONGED FASTING. 

"I felt very bright this morning because the sun was shining very brightly 
and the day is so fine and so sunny that it looks very much like one of my 
Maltese days for which I am very desirous. This afternoon I have been on 
the roof of the laboratory so as to enjoy the sun, which I love so much because 
I was born and bred in it, and I do not feel very fatigued at all or any appre- 
hension or any fear. I feel very hopeful that my fast will be not only a 
very long one but a very successful one, and I feel no pangs of hunger whatever 
or any other sensations in my stomach, and I think that my health is progress- 
ing because the catarrh of the head in my nose and in my pharynx before the 
starting of the fast has nearly disappeared and I can aspirate very easily 
through my nostrils, a thing which I could not do very easily before, and also 
I feel that my eyes are getting less congested and better. I do not feel any 
dizziness at all, and I have never felt it, neither when I lean down to get up 
something that falls from the hand; in fact, I can come down the steep stairs 
of my balcony apartment very easily without taking hold of the railing and 
without any necessity of being helped by any one. My legs feel very strong 
and I do not feel any numbness in them." 

The subject went from the office to the calorimeter laboratory on the elevator 
and drank a glass of water, but did not return to the balcony, as it was almost 
time for the psychological tests. The rectal thermometer was used during 
the bed calorimeter experiment, observations being made every 10 minutes 
during the night. 

April 22, 1912 {ninth day of fast). — On opening the calorimeter chamber no 
odor was noticeable. The subject reported a very comfortable night and said 
that he felt very well, except that he had had a bad taste in his mouth for 
several days. He slept very quietly, ringing the bell only three times. 
During the day he had visits from a number of scientific men, with whom he 
talked considerably. He was particularly interesting and vivacious in dis- 
cussing his experiments with Professors Cannon and Folin and with Dr. 
Cathcart. He reported that he did not now feel so much inclined to mental 
work, explaining this fact on the ground that "the brain was occupied in 
eliminating the impurities from the blood." 

April 23, 1912 (tenth day of fast). — This morning L. said that the preceding 
night was the best that he had had in the calorimeter. He was very quiet 
during the night and rang the bell but a few times. He still complained of the 
bad taste in his mouth. 

In the afternoon a respiration experiment of two periods was made with the 
subject sitting in a chair. In this experiment certain abnormal values were 
obtained for the pulse-rate and the respiratory quotient, and Mr. Carpenter 
thought that there was a slight irregularity in the action of the heart. Since 
the man was essentially different as to his total metabolism from the day 
before, Dr. Goodall was asked to examine him during Dr. Langfeld's tests. 
Dr. Goodall noted the heart rate before, during, and after the dynamometer 
test, and pronounced the subject to be in an excellent condition. 

L. had been less self-assertive for the week previous, much more inclined 
to cooperate in the experimental routine, and a little more reconciled to the 
fact that the observers were familiar with their technique. He was more 
quiet and less argumentative than formerly and was at times considerably 
depressed, spending much of the time in thinking of home. In conversation 
he seemed very intelligent and interested in everything, and by no means 
lethargic. He professed to have no discomfort from the rectal ther- 
mometer, although he stated frankly that he feared its use very much before 
coming to America. He also was becoming accustomed to the respiration 
chamber and did not find it at all irksome. 



GENERAL HISTORY OF FASTING EXPERIMENT. 43 

During the respiration experiment in the afternoon he spoke of the delightful 
sensation that he had in these experiments, saying that the sound of the rotary 
blower did not disturb him, but on the contrary it seemed soothing. This 
effect is not unusual with other subjects of respiration experiments, the purring 
sound of the blower frequently producing drowsiness, so that it is difficult for 
the subjects to keep awake. 

April 24, 1912 (eleventh day of fast). — Although L. rang the bell but once an 
hour throughout the night, he maintained that he did not sleep well and conse- 
quently felt somewhat irritated when he came out of the apparatus. Although 
the weather was fine, he was quite depressed during much of the day. He said 
that it was the twelfth anniversary of his marriage, and he spent the whole day 
thinking of his family and his home in Malta. 

In the respiration experiment in the afternoon, with the subject sitting, 
the nose-pieces troubled him somewhat, and he became very irritable before 
the experiment was finished. He was, in fact, quite intractable for a time, 
complaining bitterly about the rapidity with which the experiments were 
made and how frequently he was passed from one observer to another. In 
discussing the matter, he said that it was not the experiments themselves which 
troubled him and he did not particularly object to the respiration experiments, 
but that when he had finished and opened his eyes, he saw the next observer 
waiting for him, and he thought the tests made upon him were not far enough 
apart. On beginning certain of the psychological tests, he stopped and 
refused to go any farther, saying that he wanted to have 15 minutes' rest. 
He was very irritable and said that irritability was bad for him. At such 
times, the best method to use with him was to say that a part of the experi- 
mental routine would be omitted. For instance, he was told that the alveolar 
air test would be omitted, but he was much distressed and insisted that it be 
made as usual. Again, the evening before, one of the assistants was directed 
not to insert the rectal thermometer unless the subject insisted upon it, but to 
say to L. that we did not wish to tire him with too many observations. The 
subject was much disturbed by this omission and insisted that the thermometer 
should be inserted. He was frequently inconsistent, at one time saying that 
there were too many experiments, too much going on, too many people about 
him, too many questions asked as to his feelings, and then almost immediately 
he would complain of being left too much alone to brood and to think of his 
family. He complained bitterly of his inability to get out of doors. He 
enjoyed very much his stay on the roof on April 21, but the weather had been 
raw and cold and, as he seemed very sensitive to cold, it did not seem wise to 
give him an outing. 

April 25, 1912 (twelfth day of fast). — The records of the observer showed 
that L. slept very well the night before, as the subject rang the bell only once 
or twice throughout the whole night. He reported himself as being very 
comfortable. To lessen the number of observations so as not to overtire him, 
the tests of the blood-pressure and the alveolar air were omitted. In the 
afternoon L. went for a ride in a closed carriage; this he seemed to enjoy very 
much indeed. On this day, also, Dr. Southard examined him again, having a 
long talk with him. A letter which he received from his wife stated that his 
brother-in-law was very ill and later in the day he read in a Maltese paper of 
his brother-in-law's death. This depressed him very much. 

April 26, 1912 (thirteenth day of fast). — L. slept fairly well in the calorimeter, 
ringing the bell about once an hour throughout the night. No odor was 
apparent when the chamber was opened. L. said that he was very comfort- 
able; he seemed bright and in good spirits. In the afternoon he again went 
up on the roof, and subsequently a respiration experiment, with two periods, 



44 A STUDY OF PROLONGED FASTING. 

was made with him while he was sitting in a chair. He was very drowsy and 
thought he could sleep throughout the whole night, as the previous night he 
had not slept as well as formerly. As he does not read in the evening, he found 
it somewhat tedious to sit idle until 8 o'clock, when he usually went into the 
calorimeter, and he asked to go into the apparatus an hour earlier. A respira- 
tion experiment was made early in the evening just before he went into the 
calorimeter, and this experiment was made a part of the experimental routine 
on subsequent days. He entered the bed calorimeter at 9 p. m. 

April 27, 1912 {fourteenth day of fast). — The subject reported that he had 
had a very comfortable night; he rang the bell but twice during the whole time 
he was in the calorimeter chamber. About 10 h 45 m a. m., L. was taken in the 
elevator to the third floor of the laboratory, where he was shown about and 
several pieces of apparatus were explained to him. He then walked down- 
stairs to the second floor and when about a third of the way down, he com- 
menced to move rapidly and finally actually ran down the stairs. In the office 
he looked at some photographs and dictated the following statement: 

"To-day is the fourteenth day of my fast. I feel exceedingly well; I feel 
cheerful and hopeful of the grand success. I have slept several good hours 
during the night, and am enjoying with great enthusiasm all the experiences 
that are carried on me by the professor. I have to-day with Professor 
Benedict gone around the upper floors of the laboratory and I have admired 
all the great formalities that I have seen." 

Subsequently he walked out into the hall and down-stairs to the calorimeter 
room, where he sat down in an arm chair and was the subject of a respiration 
experiment. 

In the afternoon I spent about an hour with L. on the balcony. He was very 
cheerful and talked a great deal, gesticulating freely with his hands. He 
seemed to enjoy talking about the "beautiful island of Malta." He wore 
very much lighter clothing to-day, not using his heavy blanket wrapper at all. 
In fact, while walking about the laboratory, he carried his blanket wrapper 
over his arm, and although the office was a little cool, he did not put it on, but 
sat near the window. 

One of the striking features of this fast is the fact that it has been so different 
from what he expected in many ways, a good illustration of the unreliability 
of personal impressions. For instance, he had expected to feel very chilly, 
but while he said he felt chilly the first two or three days, it is possible that 
this may have been due to his imagination, as subsequently he was not nearly 
so chilly and on this day asked that the temperature of the room should be 
lowered, which was done. He also said one of the first days that he was in 
the laboratory that he would have a large amount of phlegm as the fast pro- 
gressed, but this had not been the case up to this date. While talking with 
Dr. Ash on this day, L. said that he had expected that he would become very 
hoarse during the fast, but thus far his voice had not changed, except to grow 
clearer. In the evening he was given a bath, his body being sponged with 
distilled water, after which he entered the calorimeter. 

A )ril 28, 1912 {fifteenth day of fast). — The experimental records show that 
the subject slept very well throughout the night and he himself reported a 
good night. In the afternoon he was taken out for a carriage drive, with the 
windows of the carriage open. He was well wrapped up, and said that he was 
perfectly comfortable and did not get too tired. He walked down to the 
basement and out to the carriage and on his return walked from the carriage 
up-stairs again into the calorimeter laboratory. He then sat down for several 
minutes and later a photograph was taken of him, with the blackboard showing 



GENERAL HISTORY OF FASTING EXPERIMENT. 45 

his body-weight curve as a background. He seemed unusually strong and 
active, walked with certain feet, and did not seem to be in any way exhausted 
or tired. He talked very excitedly during the whole of the drive; in fact, he 
talked continuously. He said that about 6 o'clock that morning he had had 
a normal seminal emission, which did not irritate him as did the one just prior 
to the fast. His observation was verified by Dr. Goodall's subsequent exami- 
nation of the urine, a large number of spermatozoa being found. 

April 29, 1912 {sixteenth day of fast). — When the calorimeter was opened in 
the morning, the usual absence of odor was recorded, and the subject reported 
himself as very comfortable and as having slept very well. Inasmuch as it 
was not L.'s custom to use a toothbrush — at least no toothbrush was found in 
his luggage and he had not brushed his teeth since his arrival in Boston — it 
seemed desirable to obtain some cultures of the micro-organisms of the mouth. 
These were secured by Dr. Kendall, of the Harvard Medical School, but the 
examination of them showed only adventitious organisms. L. was much inter- 
ested in this test, especially as such a test had never been made in a previous 
fast. Early in the forenoon, he was very irritable and complained of the exces- 
sive amount of drinking water; he was especially disturbed because he had not 
heard from home. While Mr. Carpenter was making the respiration experi- 
ment in the afternoon, the subject did not seem quite so well as on other days. 

April SO, 1912 (seventeenth day of fast). — When he came out of the chamber 
in the morning, L. reported that the night had been one of the best he had 
had since coming to the laboratory. The observer said that there was no odor 
and that the subject was very comfortable. Aside from the taking of a series 
of new photographs, the day was uneventful. 

May 1, 1912 (eighteenth day of fast). — The observers' records show that the 
subject scarcely moved during the night, and that the bell rang only once or 
twice. There was no odor when the apparatus was opened. In the afternoon 
of this day, L. was taken for a drive. 

May 2, 1912 (nineteenth day of fast). — L. reported a very comfortable night 
in the calorimeter; no odor was apparent when the apparatus was opened. 
At this stage of the fast, it had become very clear to the observers that L. had 
changed since the beginning. While he walked slowly and steadily and gave 
no special evidence of weakness, he was much less talkative, less inclined to 
offer advice, and more quiet and subdued. He was by no means so active a 
man as at the beginning of the fast. He complained that for several mornings 
past he had "felt his bones" during the night and was so uncomfortable that 
he waked up and turned over. It had become increasingly difficult to adjust 
the stethoscope so as to obtain a clear record of the pulse-rate. On this day 
so much difficulty was experienced that an assistant was detailed to watch the 
movement of an artery in the neck. His voice was somewhat weaker, though 
not husky, but it occasionally rose to its original tone. Furthermore, although 
he had maintained that about this period of the fast he would be intellectually 
better, he was in reality disinclined to read, write, or talk as much as at the 
beginning. This was well brought out by the fact that when he was told on 
this day that he ought to spend his time studying the principles of the appa- 
ratus and the technique rather than to trouble himself about the results, he 
stated that he did not want to spend so much time in this way and that he did 
not find himself very much interested in them, but that he would be better 
later on. This appears to contradict the statement that he would be more 
active intellectually as the fast progressed. He was measured by Dr. Anderson 
on this day, who remarked that, as he saw him only once a week, it was very 
obvious to him that he had changed very much since the beginning of the fast. 
His loss in flesh was of course much more apparent to Dr. Anderson than to 



46 A STUDY OF PROLONGED FASTING. 

those of us who saw him daily. In the afternoon he was taken to the roof 
again, where he stayed for a little more than an hour. When he returned to 
the calorimeter laboratory, he sat down in a chair until the psychological tests 
began. He still had a great interest in the various tests and made the sug- 
gestion that the rectal thermometer be inserted before the evening respiration 
experiment, so as to get the temperature changes during this experiment. 

May S, 1912 {twentieth day of fast). — Although the subject reported that he 
did not sleep so well as he did the night before, he nevertheless had a fairly 
comfortable night. There was no odor apparent on opening the chamber. 
As the weather was exceptionally fine, L. was taken for a drive in the afternoon. 

May 4, 1912 {twenty-first day of fast). — The subject reported this morning 
that he had had "several good hours of sleep." He was very comfortable, 
except that he had to change his position several times during the night. A 
number of unsuccessful attempts were made to take the pulse-rate with the 
string galvanometer, but the apparatus was broken. In the afternoon the 
subject was taken out driving. In the evening much time was spent in finding 
a suitable location for the stethoscope, and when the place was finally found 
it was so sharply localized that a movement of the stethoscope bell I cm. in 
any direction would pi event the taking of good records. It was then agreed 
upon that if during the night the pulse records were not obtainable, a signal 
to L. would be given, who would place the bell of the stethoscope as nearly as 
he could in the place decided on. Fortunately, on this and also on the fol- 
lowing night the pulse-rate was secured with considerable regularity. 

May 5, 1912 {twenty-second day of fast). — The experimental records showed 
that the subject remained very quiet throughout the night, ringing the bell 
only once or twice. In the afternoon of this day he was taken out for a carriage 
drive of two hours. He enjoyed this drive very much indeed, speaking enthu- 
siastically about the beauty of Boston and its suburbs, and how it elevated him 
to see it. When he returned, he claimed that he was neither tired nor cold, 
although the air was cooler than when he went out the day before. 

He had changed considerably during the preceding week, a fact which was 
noted by several persons who had not seen him during that period. He 
walked much more deliberately, but appeared to be perfectly sure of his footing. 
When he came down-stairs to take the psychological tests, he took the last 
three or four steps quite rapidly and stepped off briskly to greet Dr. Langfeld. 
On the other hand, at the end of the psychological test, he sat rather dejectedly 
in his chair. When Dr. Langfeld told him that the tests were over, however, 
he rose from the chair immediately, and returned to the balcony, but stopped 
half-way up the stairs to look at the string galvanometer with which we were 
working, and seemed as intellectually keen and bright as any one. 

May 6, 1912 {twenty-third day of fast). — On coming out of the chamber, L. 
reported that he had a fairly comfortable night, sleeping very well. He com- 
plained, however, that as he had lost so much adipose tissue, he found it 
rather difficult for him to sleep on one side for any length of time, and he was 
obliged to turn from one side to the other frequently. The observers also 
found it extremely difficult to get the records of the pulse-rate. L. was much 
stimulated as a result of visitors in the forenoon. He exhibited a much 
sharper intellectual activity than had been apparent for several days, which 
might possibly prove his statement that he would become more intellectually 
keen as the fast progressed. Other than that we could see little evidence of 
the so-called intellectual activity that he had referred to frequently during 
the fast. 

May 7, 1912 {twenty-fourth day of fast). — The subject was fairly quiet 
throughout the night, moving only occasionally. When the calorimeter was 



GENERAL HISTORY OF FASTING EXPERIMENT. 47 

opened, no odor was observed. Thus far in the fast, no acetone odors had 
been noted. In the forenoon a number of visitors came into the laboratory 
and several small photographs were taken of L. In the afternoon respiration 
experiment, when the subject sat writing, he was reported as being very 
obstinate and intractable. He seemed to have some difficulty in finding a 
position to suit him, and the nose-pieces had to be inserted a second time to 
make sure that they were well fitted. He also complained that the writing 
paper was too heavy, so that he would have to write on both sides of the paper 
to save postage. He appeared to be very difficult to please. At 5 o'clock on 
this day, L. reported that he was feeling very much depressed, owing to the 
weather, and that he spent the whole time thinking about home and was 
much worried. 

May 8, 1912 (twenty-fifth day of fast). — The subject reported a veiy comfort- 
able night, moving only once or twice during the night. In order to control 
the possible influence of small muscular movements on the temperature of the 
air in the chamber at the end of an experimental period, the height of the 
spirometer attached to the calorimeter was recorded on a smoked paper drum, 
a routine which proved very helpful on subsequent nights. On this morning 
L. was very sullen and disagreeable when spoken to, especially with Mr. 
Carpenter. He said that he had expected there would be a yellow pigmen- 
tation of the skin as in other fasts, but this did not appear. In the afternoon 
I asked him how much longer he wished to fast. He replied by asking how 
long I wanted him to fast. As I had previously told him on several occasions 
that I should like him to fast 30 days, it was obvious that he wished to be 
asked to stop fasting rather than to break the fast himself. I told him that 
we should probably wish him to fast for 30 or 31 days. The diet of the 
re-alimentation period was then discussed. He believed very strongly that 
the fast should be broken with acid fruit juices alone. The rectal thermometer 
remained in position throughout the day and the temperature records were 
secured for the greater part of the time. 

May 9, 1912 (twenty-sixth day of fast). — The subject spent an unusually 
quiet night, ringing the beU only three times throughout the night. In the 
afternoon, during an active conversation which I had with him, in which 
general questions were discussed, he seemed to be very spirited, lively, and 
interested. His pulse-rate was 86 at this time. In the evening, when he came 
down-stairs to take the psychological test, he appeared to be quite unsteady 
on his feet and said that he was light-headed. He thought that his unsteadi- 
ness was due to the fact that he had lost so much weight that he put unusual 
strength on his foot, more than he needed to take his weight, and that this 
tended to unbalance him. Mr. Carpenter also noticed that when walking, L. 
was quite uncertain in his steps and thereafter we watched him more closely 
when he was on his feet. He said considerable on this day about the method 
of breaking the fast, emphasizing the fact that he wished to be able to say that 
the fast was broken by request and not by his desire. It was arranged to 
allow him to fast for 31 days, and then to begin taking food. He also expressed 
a preference for breaking his fast on a diet consisting of boiled rice and honey, 
with the juice of oranges and lemons and grape juice. Obviously such a diet 
should have many scientific points of interest, but it was questionable 
whether, after fasting so long, the stomach should be filled with the acids of 
oranges, lemons, and grapes. He was firmly convinced that this was the diet 
that should be used and could not be dissuaded from it. Subsequent experi- 
ence proved the undesirability of this kind of a diet for breaking a long fast. 

May 10, 1912 (twenty-seventh day of fast). — The subject reported that he did 
not sleep so well during the night as he had the night previous. He was not 



48 A STUDY OF PROLONGED FASTING. 

uncomfortable in any way, but simply could not sleep. There was no notice- 
able odor. In a discussion with some physicians who visited him to-day, L. 
emphasized the fact that his tongue remained coated throughout the whole 
fast and that he had more or less of a bad taste in his mouth. His theory was 
that the waste products were not all eliminated from the body and that "they 
were trying to find their way out by the mouth." He thought that if he had 
defecated he would have had less discomfort. Although the subject went for 
a drive on the afternoon of this day, he complained that the weather had 
changed and that it was not pleasant. He also complained bitterly to Dr. 
Langfeld regarding Mr. Carpenter, saying that he would like to break every 
bone in his body. This would pronounce against fasting for amiability. 

May 11, 1912 {twenty-eighth day of fast). — L. reported that it was late before 
he fell asleep and that he had not been able to sleep much or soundly during 
the night. In the early part of the evening, after he had been sealed into the 
calorimeter, he signaled that he was obliged to urinate and he had to be taken 
out. When he was sealed up again shortly after 10 h 30 m p. m., he thought he 
did not go to sleep until 12 o'clock. He moved several times during the night. 
During the afternoon of this day he spent some time on the roof, reading, and 
seemed to enjoy himself very much, appearing to be in good spirits all of the 
afternoon. But when he came down from the roof, he was extremely stubborn 
when taking his tests with Dr. Langfeld. For example, in a list of ten words 
which was read to him for memorizing, the eighth word was "wart." L. did 
not understand what it meant and asked to have it explained. Dr. Langfeld 
tried to keep him quiet until the list was read a second time, but he would not 
listen and insisted upon knowing what the word was. When Dr. Langfeld 
read the list to him a second time, he said that he had not heard the words, as 
he was thinking of what the word "wart" meant. 

May 12, 1912 (twenty-ninth day of fast). — The subject said that he slept very 
well the previous night and in the morning seemed very bright. He busied 
himself much of the forenoon in writing a history of his life. (See p. 22.) 
A respiration experiment was made with him in the morning in which the 
subject breathed an atmosphere containing a large percentage of oxygen. 
In the afternoon I took him for a drive of 2 hours. Duiing the drive he 
was bright and very talkative, telling a good deal about his family, his per- 
sonal history, and his difficulty in getting an education. He became more 
quiet before he returned to the laboratory and in the psychological test he 
told Dr. Langfeld that the weather depressed him very much. He also 
told Dr. Langfeld that he was very sorry that I wanted him to break the fast 
and that he could easily fast for 10 days more; he did not like to break his 
fast until his tongue had become clear. On the other hand, he said that he 
would be out of the laboratory on May 23 and would be a free man again and 
"have free ice cream." It was evident that he would be difficult to control. 
There has been none of the pigmentation of the skin which he had expected 
and very little phlegm. 

May 18, 1912 (thirtieth day of fast). — L. reported that he had slept well. He 
was in excellent spirits on this day and very lively and jolly. His movements 
were vigorous and he was unusually bright and active. All of his spare time 
was occupied in writing his autobiographical notes, which at this time covered 
13 closely written pages. As usual he felt quite able to be photographed. 
At noon I asked him how he was and he replied that he felt very well indeed, 
saying: "We have seen the lighthouse and are now entering the harbor and 
will see land shortly." 

In the afternoon I took him to the Boston City Hospital, where Dr. Francis 
H. Williams made an extensive series of X-ray photographs. L. was intensely 



GENERAL HISTORY OF FASTING EXPERIMENT. 49 

interestedin this, and also thoroughly enjoyed the ride to and from the hospital. 
He was very much pleased with the whole day's program. He said that he 
was "now dropping anchor and therefore at the end of the voyage." His 
mental condition seemed to make a great difference in his whole make-up. 
On some days his faculties were veiy much keener than on others. For in- 
stance, during the test for visual acuity, he answered the position of the 
letter "E" with great strength of voice and with no hesitation whatever. 

May 14, 1912 {thirty-first day of fast). — On this day, which was the last day 
of the fast, the program was very full, including an extensive series of photo- 
graphs (see Plate 5) and the body measurements. The first word that L. 
spoke when the calorimeter was opened was in regard to having his photograph 
taken. In response to an invitation issued by the Nutrition Laboratory, a 
number of medical men from the vicinity of Boston came to see the subject 
between 2 and 3 o'clock in the afternoon. L. talked very rapidly and in a 
lively manner for nearly 40 minutes, setting forth his views in a more or less 
lucid manner to his audience. In the middle of his talk, I took his pulse-rate and 
found it to be 82. The sub j ect was particularly desirous of having his photograph 
taken during the afternoon and wished to be photographed with each individual 
who had worked with him. In this afternoon talk with the physicians, L. 
made many conflicting statements. For instance, he said that he never used 
wine or alcohol in any form, while he had repeatedly told me that he frequently 
drank Malta wine and was not a teetotaler. He also reported that he was 
invariably a vegetarian, fruitarian, and nutarian, eating no meat; this was 
strikingly in contrast with the fact that he ate meat several times in the days 
just preceding the fast. During the visit of the physicians, he became very 
much excited and enthusiastic, evidently enjoying the opportunity of speaking 
to his audience. During the respiration experiment in the evening, he talked 
considerably about the end of the fast, saying that while he was not exactly 
tired, yet there was much emotion connected with his fast and he was thinking 
of the effect produced at home when the news reached them that he had 
completed the fast of 31 days successfully. Malta was awaiting the news and 
all of the people would be discussing his wonderful feat. 

Figures D, E, and F in Plates 2 and 3 (pages 19 and 31) are from photo- 
graphs secured on this last day of fasting. Figure E is of special interest, as 
it shows L. posing for 20 seconds while the exposure was being made. There 
is no evidence of unsteadiness. 

RE-ALIMENTATION PERIOD. 

May 15, 1912 (first day with food). — The bed calorimeter experiment was 
ended at 7 h 56 m a. m. When the apparatus was opened there was no odor 
apparent. The observer reported that the subject was comparatively quiet 
throughout the night, with but one movement between ll h 44 m p. m. and 
5 h 28 m a. m. L. complained, however, that he was kept awake by acid fumes 
and noise in the calorimeter room. Photographs were taken of the subject 
while he was upon the respiration apparatus, and also later in various positions. 
When he was weighed after the respiration experiment was over, it was found 
that he had lost 13.25 kilograms of his initial weight. He seemed much 
pleased that the fast was ended and spoke of the excitement there would be at 
his home when the cablegram was received announcing the successful com- 
pletion of his fast. 

The method of breaking the fast had been thoroughly discussed with the 
subject previously, but he insisted upon using lemons, oranges, grape juice, 
and honey. It was finally arranged that he should follow his own choice in 
the matter and, accordingly, lemons, oranges, boiled rice, and honey were 



50 A STUDY OF PROLONGED FASTING. 

supplied him at his request. At 9 h 38 m a. m. he peeled and ate two lemons, 
using neither water nor sugar. In addition, three oranges, about 300 grams 
of honey, and approximately 1 liter of grape juice were taken in portions 
during the day. To study the flora of the colon, a rectal injection of sterile 
salt water was given him at l h 10 m p. m. About 4 h 30 m p. m. he complained of 
severe colic, but was somewhat relieved about 5 h 30 m p. m. by a copious move- 
ment of the bowels. Subsequently he again had colic, with a second move- 
ment of the bowels and vomiting. He appeared to be utterly wretched and 
weak, and to have entirely lost his courage. The contrast between his high 
spirits of the day before and his condition on this day was very striking. 

No respiration experiments were attempted on this day after the ingestion 
of food, but a sample of the blood was taken, and records were made of the 
blood pressure and the alveolar air. Under the circumstances, it did not seem 
wise to have the subject sleep in the calorimeter. Arrangements were there- 
fore made for him to sleep on a couch in the calorimeter laboratory, with a 
physician in constant attendance. While no records of the metabolism were 
obtained during the night, the pulse-rate was recorded by means of the 
stethoscope, the assistant sitting behind a screen out of sight of the patient. 

May 16, 1912 (second day with food). — The subject passed a restless night, 
having another attack of colic about 2 a. m. and defecating. The pulse-rate 
through the night was somewhat higher than on previous nights. As he 
seemed weak and sick in the morning, Dr. Goodall, who came to see him early 
in the day, urged him to take some weak beef tea or clam broth, also toast, 
but he utterly refused to take anything of this nature, saying that such food 
would poison him. He furthermore said that as we were now through with 
him, we wished to poison him. He was as firmly convinced as ever that his 
method of breaking the fast was the correct one, but thought that he took food 
too soon and that he should have waited until his tongue had cleared, as he 
considered it dangerous to break the fast sooner. He told the physician in 
attendance during the night that a man might have to fast 100 days before 
the natural hunger would return. L. lay on the couch until about 10 a. m., 
when he went to the balcony and dressed. He continued his diet of fruit juices 
and honey throughout the day, but, at our urgent request, diluted the lemon 
juice with distilled water, taking first 77 c.c. of lemon juice, an equal amount 
of distilled water, and 11 grams of honey. Between 9 h 30 m a. m. and 7 h 30 m 
p. m., he drank at intervals a mixture containing the juice of 6 oranges, 367 c.c. 
of water, and 128 grams of honey, making a total volume of about 1,400 c.c. 
About 7 h 30 m p. m., he took the juice of another orange, saying that he was 
hungry but felt very well. During the day Dr. Goodall made the usual 
physical examination; the blood pressure and the alveolar air were also 
observed. The subject entered the bed calorimeter at 8 h 14 m p. m. for the 
usual night experiment. 

May 17, 1912 (third day with food). — At 10 p. m. the subject telephoned and 
urinated, but after that time the experimental records show that he was very 
quiet. When he left the apparatus in the morning, he said that he had had a 
very comfortable night, sleeping better than he had for years. He thought 
he must have had at least 10 hours of dreamless sleep. He appeared to be in 
good spirits. The usual respiration experiment was made with him this 
morning. Throughout the day, he continued taking at intervals a mixture 
of orange juice and honey, diluted with water; he was also supplied with 
raisins and dates. The usual physical examination was made by Dr. Goodall; 
also tests of the blood pressure, the alveolar air, and for acetone in the breath. 
A sample of the blood was taken by Dr. Ash. In the afternoon he went up on 
the roof for a time, but was ill-humored, very difficult to please, and full of 



GENERAL HISTORY OF FASTING EXPERIMENT. 51 

complaints. The day before I had taken exception to his statements that 
we were to blame for his illness on taking food and had shown him that he 
had chosen his own diet and had refused to be guided by the advice of the 
physicians. He was greatly offended at this, and was disagreeable to every 
one in consequence all of this day. When Dr. Langfeld came at about 5 
o'clock for the psychological tests, he burst forth in a long tirade, complaining 
of nearly every member of the staff, and of his treatment at the laboratory. 
He claimed that throughout the night, while in the calorimeter, he was left in 
charge of "boys," who paid no attention to him, and that he might die there 
without any one knowing of it. He had said nothing about this to me, but 
asked Dr. Langfeld to tell me. 

In the evening L. again had colic, with diarrhea and vomiting. He entered 
the calorimeter for the regular experiment at 8 h 35 m p. m., but only on the 
condition that Mr. Emmes or Mr. Carpenter should stay in the calorimeter 
room all of the night. Accordingly Mr. Carpenter slept on a couch in the 
balcony, so as to be near if needed during the night, the usual experienced 
observers being on duty. At 10 p. m. it was reported to me that the subject 
had a pulse-rate of 104 and body-temperature (rectal) of approximately 38° C. 
I went to the calorimeter laboratory and remained there for a time, thinking 
that he might be suffering from colic and would have to be taken out. About 
HP 30 m p. m. he telephoned that he wished to defecate. He was accordingly 
taken out of the calorimeter. After defecating, he returned to the apparatus 
and the experiment began at ll h 57 m p. m. 

May 18, 1912 {fourth day with food). — The subject had a restless night, 
being evidently suspicious, unhappy, and discontented. He called for Mr. 
Carpenter about 3 a. m. and asked if something had not fallen upon the calo- 
rimeter, as he had felt a jar. He was assured that nothing of the kind had 
happened, both of the observers having been sitting quietly at their respective 
posts when he called. He was taken out of the apparatus at 5 h 45 m a. m. and 
said that he had not slept all night. His pulse-rate and respiration-rate were 
high ; the body-temperature was also slightly elevated. He seemed apprehen- 
sive and was apparently mentally unbalanced, talking in an irrational way with 
Dr. Goodall. The usual respiration experiment was made with him on this 
morning and he was weighed, but no other observations were taken. 

After the respiration experiment, he complained of exhaustion, but finally 
returned to the balcony and began to eat; his attitude towards nearly every 
member of the staff was suspicious and antagonistic. He insisted that the 
British consul should be asked to come to the laboratory to confer with him, 
which was done. It was finally decided that considering his condition and 
general attitude, he would more quickly recover his balance if cared for at a 
hospital. Arrangements were made with the Massachusetts General Hospital 
to defray the expense of his care and he was taken there in a carriage by one 
of the members of the staff, and placed in a private room. On May 19 he 
was seen by Dr. Goodall, who found him very contrite. He asked Dr. 
Goodall to find me at once and to apologize for his behavior the day before. 

By permission, a copy of the hospital records is appended herewith, giving 
the history of his stay in the Massachusetts General Hospital. 

"May 18. — Patient states that he completed a 30-day fast under observation 
at the Carnegie Nutrition Laboratory three days ago, and that he was sent 
here to-day through a misunderstanding and that he should not be in a 
hospital. History and physical examination not attempted by order of Dr. 
Langnecker, superintendent on duty. 

"May 19. — Seems contented, cheerful, and comfortable. Eats with good 
appetite and relish. Says the bed is not as comfortable as the calorimeter. 



52 A STUDY OF PROLONGED FASTING. 

"May 20. — Patient was very angry on coming into the hospital, first, 
because he was in a public institution, whereas he had expected to go into a 
'convalescent home for wealthy people.' Second, because the fact of his 
going to a hospital after his fast would detract from the renown and interest 
of his feat. 

" Yesterday he appeared very happy. He seemed to realize and accept that 
there were certain disadvantages here. He was asked what diet he wanted 
and requested straight house diet, and this was given him in accordance with 
the ideas of Dr. Benedict. He ate this with relish. He appeared a little weak, 
but able to be about. He spent the day in writing and in sorting out various 
articles and pictures, of which he had a half of a valise full. 

"Last night he was nauseated and vomited several times, but was not ill 
enough to be reported. This morning he was again very angry. He com- 
plained that oatmeal was very bad for a man recovering from starvation. 
He repeated his complaints of Saturday (May 18) and said that he was being 
held by conspiracy and wanted to summon or go to the British consul. He 
was therefore discharged at his own request. 

"While at the front door waiting to have his bill arranged and his valuables 
procured, he complained loudly that his possessions were being withheld from 
him. While waiting for a carriage to be summoned, he called out loudly many 
times that he would get out of here if he had to crawl on his hands and knees. 

"While in the wards he was shown every attention by the nurses, and 
especially all food and drink brought to him as desired. 

"Discharged to own M. D. (Not treated)." 

Subsequent to his leaving the Massachusetts General Hospital, L. was seen 
by Drs. Langfeld and Goodall and a second set of X-ray photographs was 
taken by Dr. Williams. Such data as were secured are included in the 
individual reports. 



PHYSICAL CONDITION OF THE SUBJECT DURING THE FAST. 

By Harry W. Goodall, 
Department of Physiological Chemistry, Harvard Medical School. 

A complete physical examination of the subject was made 60 hours 
before the beginning of the fast, on the first day of the fast, and every 
alternate day thereafter. So far as possible, the same conditions as to 
time of day, posture, methods, etc., were observed at each examination. 
The percussion outline of the various organs was marked with pencil and 
these lines were not disturbed until variations in the size of the organs 
made it necessary. Owing to the thinness of the subcutaneous tissue 
the percussion note could be sharply defined and the favorable mental 
attitude of the subject permitted complete muscular relaxation. 

Family history. — Father 68 years of age, mother 64 years, both living and 
well. Two sisters, one 27 years of age, the other 19 years, both living and 
well. One sister died in infancy of malaria. One brother died in childhood 
of croup. Twin brother died in infancy, cause unknown. 

Past history. — Was a delicate child, the most pronounced characteristic 
being a sensitive nervous organization. Does not remember about diseases 
of childhood. In his twentieth year (20 years ago), while in the university, 
he had a severe nervous breakdown and had to give up his studies. Has 
suffered from neurasthenia since. Has been under the care of several physi- 
cians at different periods without receiving any benefit. Being discouraged 
with these experiences, he became interested in certain popular articles advo- 
cating fasting as a cure for diseases, including neurasthenia, and in April 1910 
he underwent his first fast. So great was the improvement in his nervous 
condition following this fast that he has since devoted much of his attention 
to the study of fasting. Previous to the present experiment he has undergone 
the following fasts: 

April 1910. Fasted 8 days, taking nothing but water; not under obser- 
vation. 

March 1911. Fasted 40 days, taking nothing but water; under partial 
observation. 

August 1911. Fasted 12 days, taking nothing but water; under partial 
observation. 

November 1911. Fasted 5 days, complete fast; not under observation. 

These experiences have convinced him that, in health, food not only 
makes the individual susceptible to diseases and causes disease, but 
also interferes with the proper exercise of the mental faculties. He 
states that during the fasting period the mind is clear and alert and 
that there is a strong desire for study. It is now the practice of his 
wife, his two children, and himself to abstain entirely from food during 
any illness, and he is convinced that the severity of any disease is 
reduced, the subjective symptoms made less disagreeable, and the course 
of the disease shortened if such a course is pursued. 

53 



54 A STUDY OF PROLONGED FASTING. 

His personal experiences and his study have been so convincing that 
he was desirous of undergoing a scientific experiment, under the most 
perfect conditions, for the benefit of humanity. Having learned that 
the most perfect equipment in the world was at the Nutrition Labor- 
atory, he applied for the privilege of undergoing this experiment there. 
He expressed his pleasure at the cordial manner in which he had been 
received at the laboratory, and stated that he was very happy in the 
thought that he was about to undertake the fast, not alone on account 
of the scientific value of the experiment, but also because of the im- 
provement which he anticipated in his own well-being. He expressed 
his pleasure in going into minute details as to his subjective feelings. 

RESULTS OF PHYSICAL EXAMINATION. 

April 11, 1912 (60 hours before beginning the fast): 

A well-developed, well-proportioned, and fairly well-nourished man. 
Height, 170.7 cm.; weight, 60.1 kilograms; age, 40 years. Stands 
erect. Walks with body erect, with no abnormalities in the gait. 
Hair of head and beard dark. Skin has a muddy yellowish tinge, but 
is soft and moist. Slight conjunctivitis of both eyes. Very moderate 
amount of subcutaneous fat. Muscles of moderate size but rather 
soft. Has a small infected papule on left alae nasse. No pulsations 
noted in the neck, chest, or abdomen. No visible abnormalities. 

Mouth: Mucous membrane of the lips and cheeks of good color and moist. 
Tongue moist with a slight coating, especially on the central and poste- 
rior portions. Teeth in fair condition. There is a slight deposit of 
discolored tartar at the base of the teeth and two teeth have temporary 
soft fillings. No particular odor to the breath. No enlargement of 
the tonsils. Pharynx is reddened, the blood vessels dilated, and there 
are a few blebs on the posterior pharyngeal wall with a little mucus 
adhering. 

Glands: Cervical, axillary, and epitrochlea glands not palpable. A few 
small glands in both groins. 

Reflexes: Pupils equal and react normally to light and distance. Abdom- 
inal, cremasteric, patella, Achilles, and plantar reflexes normal. 

Chest: Symmetrical, well formed, some sinking in of the supra- and infra- 
clavicular spaces. Good expansion with inspiration. No bulging 
of the praecordia, and apex beat of heart is not visible or palpable. 

Lungs: Percussion of the right lung shows normal resonance to the upper 
border of the fifth rib in the nipple line, to the lower border of the 
fifth rib in the axillary line, and to the eleventh rib in the back. On 
the left normal resonance to the eighth rib in the mid-axillary line 
and to the eleventh interspace in the back. Vocal fremitus is 
slightly increased and expiration slightly prolonged at the right apex, 
extending to the second rib in front and the spine of the scapula 
behind. There were no rales and the lungs were otherwise negative. 

Heart: The area of superficial cardiac dullness (light percussion) was 
measured from a perpendicular line through the mid-sternum and a 
horizontal line drawn at the level of the nipples. The upper border 
of cardiac dullness was at the third interspace. The left border of 
cardiac dullness was 9.5 cm. from the mid-sternum, 1.2 cm. inside 
the nipple. The right border of cardiac dullness was 1.2 cm. from 



PHYSICAL CONDITION OF SUBJECT DURING FAST. 55 

April 11, 1912— Continued. 

the mid-sternum. The total width of cardiac dullness was 10.7 cm. 
The cardiac sounds were somewhat distant, but of good quality and 
regular rhythm. There were no murmurs to be heard. The aortic 
and pulmonic second sounds were of equal intensity. 

Pulse: The pulses were equal, regular at the rate of 82 per minute. 
Rhythm regular, volume fair. No sclerosis of the vessels noted. 

Abdomen: The abdomen was symmetrical, rather prominent when 
standing, but flat when reclining. It was soft, tympanitic, but with 
no distension. There was no tenderness on palpation. Nothing 
abnormal was felt. 

Liver: The upper border of liver dullness was at the lower border of the 
fifth rib in the nipple line. The lower border of liver dullness was 

I cm. below the costal margin. Total width of dullness 11.5 cm. 
The edge of the liver was indistinctly palpable, soft, and without 
irregularities. 

Stomach: The measurements of the stomach were determined as accu- 
rately as possible by means of auscultatory percussion. The lines 
of measurement were the median line of the body and a horizontal 
line through a point half-way between the tip of the ensiform and the 
umbilicus. Tympany in the median line extended from the tip of the 
ensiform to a point 3.5 cm. above the umbilicus, a total distance of 

II cm. The left border of tympany extended to a point 16 cm. from 
the median line. Faint rhythmic sounds were to be heard with the 
stethoscope. There was no splashing with palpation. 

Spleen: The upper border of splenic dullness was at the eighth rib. The 

area of splenic dullness was vaguely determined as 7X5 cm. The 

spleen was not felt. 
Kidneys: Neither kidney was palpable. 
Genital organs : Aside from a long prepuce and a slight left variococele, 

the penis and testicles were normal. 
April 14, 1912 (first day of fast): 

Abdomen: Not as prominent. Flat to percussion everywhere except 

over the area of stomach tympany. 
Stomach: No rhythmic sounds heard. 
No change noted in general appearance, mouth, glands, reflexes, chest, 

lungs, heart, pulse, liver, spleen, kidneys, or genital organs. 
April 16, 1912 (third day of fast): 

Mouth : No change noted, with the exception of a pronounced odor to the 

breath. 
Reflexes : No change noted except patella reflex not as active. 
Liver: Lower border of liver dullness at costal margin. Total width of 

dullness 10.5 cm. Edge not felt. (First change noted.) 
Stomach: Tympany in the median line extends from a point 3 cm. above 

the tip of the ensiform to 7 cm. above the umbilicus. Total width 

10.5 cm. Left border of tympany 18 cm. from the median line. 

Active peristalsis with pronounced rhythmic sounds. 
No change noted in general appearance, glands, chest, lungs, heart, pulse, 

abdomen, spleen, kidneys, or genital organs. 
April 18, 1912 (fifth day of fast): 

Mouth: No change noted. Odor to breath still pronounced and tongue 

more heavily coated. 
Reflexes: Patella reflexes only obtained with reenforcement. 



56 A STUDY OF PROLONGED FASTING. 

April 18, 1912— Continued. 

Abdomen: Not as prominent while standing, retracted while reclining. 

Dull to percussion, except over the area of stomach tympany. 

Marked visible pulsation of the aorta. 
Liver: No change from note of third fasting day. 
Stomach: Tympany in the median line from a point 3.5 cm. above the 

tip of ensiform to 7.5 cm. above umbilicus. Total width, 10.5 cm. 

Left border of tympany 16 cm. from median line. Rhythmic sounds 

were audible but not marked. 
No change noted in general appearance, glands, chest, lungs, heart, pulse, 

spleen, kidneys, or genital organs. 
April 20, 1912 {seventh day of fast): 

General appearance : No change noted, except that the features are slightly 

drawn and subject moves about a little more slowly. 
Mouth : No change from note of fifth fasting day. 
Reflexes: No change in pupillary and plantar reflexes. Patella reflexes 

obtained with difficulty. Achilles reflex very slight. 
Heart: Left border of cardiac dullness 8.5 cm. and right border 1 cm. 

from mid-sternum. Total width, 9.5 cm. (first change noted). No 

change in character of sounds. 
Abdomen: No change from note of fifth fasting day. 
Liver: No change from note of third fasting day. 
Stomach : Tympany from a point 3 cm. above the tip of the ensiform to 

8 cm. above the umbilicus. Total width 9.5 cm. Left border 14 

cm. from median line. Rhythmic sounds heard. 
No change noted in glands, chest, lungs, pulse, spleen, kidneys, or genital 

organs. 
April 22, 1912 (ninth day of fast): 

General appearance: Conjunctivitis not quite so marked. Infected 

papule on the nose has disappeared. Otherwise no particular change. 
Mouth : Mucous membrane of tongue and mouth dry. Tongue slightly 

less coated. Odor to breath not so pronounced. 
Reflexes : Patella and abdominal reflexes absent. Achilles reflex obtained 

only with difficulty. Cremasteric and plantar reflexes normal. 
Heart: Left border of cardiac dullness 8 cm., right border 1 cm. from 

mid-sternum. Total width 9 cm. (second change in size of heart 

noted). No change in character of sounds. 
Abdomen: No change from note of fifth day. 
Liver : No change from note of third day. 
Stomach: Tympany from a point 4 cm. above tip of ensiform to 8 cm. 

above the umbilicus. Total width, 10.5 cm. Left border tympany 

14 cm. from median line. 
No change noted in glands, chest, lungs, pulse, spleen, kidneys, or genital 

organs. 
April 24, 1912 (eleventh day of fast): 

General appearance: Features somewhat drawn. Muscles not quite so 

firm. Skin more elastic. Seborrhea sicca of entire scalp. Walks 

without evidence of weakness. No unsteadiness when standing with 

eyes closed. 
Mouth: Mucous membrane of mouth and tongue dry. Lips dry and 

desquamating. Odor to breath less marked. 
Reflexes : No change from note of ninth day of fast, except Achilles reflex 

absent. 
Chest : Some sinking in of supra- and infra-clavicular spaces. 



PHYSICAL CONDITION OF SUBJECT DURING FAST. 57 

April 24, 1912— Continued. 

Heart : No change noted in measurements of heart from ninth day of fast. 
Faint systolic souffle heard all over the praecordia, loudest at the apex. 
Not transmitted, and not related to the respiratory murmur. Inten- 
sity not influenced by posture. Heart sounds not so distinct. 

Pulse : Volume of pulse not so good. 

Abdomen: No change from note of fifth day of fast. 

Liver: No change from note of third day of fast. 

Stomach : No change from note of ninth day of fast. Rhythmic sounds 
heard. 

No change noted in glands, lungs, spleen, kidneys, or genital organs. 
April 26, 1912 {thirteenth day of fast): 

General appearance: Features not so drawn. Conjunctivitis somewhat 
improved. Otherwise no change from note of eleventh fasting day. 

Mouth : No change from note of eleventh fasting day. 

Reflexes: No change from note of eleventh fasting day. 

Chest : No change from note of eleventh fasting day. 

Heart: No change from note of eleventh fasting day. 

Pulse: No change from note of eleventh fasting day. 

Abdomen: Markedly retracted when reclining. 

Liver: No change from note of third fasting day. 

Stomach: Tympany from a point 3 cm. above tip of ensiform to 8 cm. 
above the umbilicus. Total width 10.5 cm. Left border of tympany 
12.5 cm. from median line. Rhythmic sounds heard. 

Kidneys: Pole of right kidney just palpable. (First change noted.) 

No change noted in glands, lungs, spleen, or genital organs. 
April 28, 1912 (fifteenth day of fast): 

General appearance: Odor to breath less pronounced. Stands erect. 
Has normal gait but moves about more deliberately. Seborrhea 
sicca much improved. 

Mouth : Mucous membrane of mouth and lips moist. No desquamation 
of lips. Tongue slightly less coated. 

Reflexes: No change from note of eleventh fasting day. 

Chest: No change from note of eleventh fasting day. 

Heart: No change in measurements of heart from note of ninth fasting 
day. The systolic souffle which appeared on the eleventh fasting 
day is no longer heard. Intensity of heart sounds not increased. 

Pulse: No change noted from note of eleventh fasting day. 

Abdomen: Retracted. Dull to percussion everywhere except over the 
area of stomach tympany. No tenderness on palpation except slight 
tenderness over the pulsating aorta. The large intestines appear to 
be about the size of a thumb and can be rolled under the finger 
extending from the caecum up to the right hypochondrium and from 
the left hypochondrium down to the brim of the pelvis. There is no 
gurgling with pressure. 

Liver: The upper border of liver dullness is 1 cm. lower than at the 
beginning of the fast. Total width 9.5 cm. (second change in size). 

Stomach: Tympany from a point 8 cm. above the umbilicus. Total 
width 9.5 cm. Left border of tympany 13 cm. from the median line. 
Rhythmic sounds heard. 

Kidneys: No change from note of thirteenth fasting day. 

No change noted in glands, lungs, spleen, or genital organs. 



58 A STUDY OF PROLONGED FASTING. 

April SO, 1912 {seventeenth day of fast): 

General appearance: Features not so drawn, otherwise no change from 
note of fifteenth fasting day. 

Mouth : Desquamation of lips entirely disappeared, otherwise no changes 
from" note of fifteenth fasting day. 

Reflexes: No change from note of eleventh fasting day. 

Chest: No change from note of eleventh fasting day. 

Heart: Left border of cardiac dullness 7.5 cm., right 0.5 cm. from median 
line. Total width 8 cm. (third change in the size of heart noted.) 
No further change in quality of heart sounds. 

Pulse : No change from the note of the eleventh fasting day. 

Abdomen: No change from the note of the fifteenth fasting day. 

Liver : No change from the note of the fifteenth fasting day. 

Stomach : No change from the note of the fifteenth fasting day. 

Kidneys : No change from the note of the thirteenth fasting day. 

No change noted in glands, lungs, spleen, or genital organs. 
May 2, 1912 {nineteenth day of fast): 

General appearance : No change from note of seventeenth fasting day. 

Mouth : No change from note of seventeenth fasting day. 

Reflexes: No change from note of eleventh fasting day. 

Chest: No change from note of eleventh fasting day. 

Heart: Left border of cardiac dullness 7 cm. from the median line, right 
border at median line. Total width, 7 cm. (fourth change in meas- 
urements of heart noted). No change in quality of sounds from the 
note of the fifteenth fasting day. 

Pulse : No change from note of eleventh fasting day. 

Abdomen: No change from note of fifteenth fasting day. 

Liver: No change from note of the fifteenth fasting day. 

Stomach: Tympany from a point 1.5 cm. above tip of the ensiform to 8.5 
cm. above the umbilicus. Total width, 8.5 cm. Left border of tym- 
pany 11 cm. from median line. No rhythmic sounds heard. 

Kidneys: No change from note of the thiiteenth fasting day. 

No change noted in glands, lungs, spleen, or genital organs. 
May 4, 1912 {twenty-first fasting day): 

General appearance: The soft parts of the extremities not so firm and 
not so large. Otherwise no change from the note of the seventeenth 
fasting day. 

Mouth : No change from the note of the seventeenth fasting day. 

Reflexes: No change from the note of the eleventh fasting day. 

Chest: No change from the note of the eleventh fasting day. 

Heart: No change from the note of the nineteenth fasting day. 

Pulse: No change from note of the eleventh fasting day. 

Abdomen: No change from the note of the fifteenth fasting day. 

Liver: Upper border of liver dullness at sixth rib, lower border at costal 
margin. Total width 9.0 cm. (third change noted in size of liver). 

Stomach : Area of tympany the same as noted on the nineteenth fasting 
day. Rhythmic sounds heard. 

Kidneys: Right kidney palpable, pole of left kidney (for first time) just 
felt with deep inspiration. 

No change noted in glands, lungs, spleen, or genital organs. 
May 6, 1912 {twenty-third day of fast): 

General appearance : No change from note of the twenty-first fasting day. 

Mouth: Mucous membrane moist. Very slight coating on tongue. 
Very little odor to breath. 



PHYSICAL CONDITION OF SUBJECT DURING FAST. 59 

May 6, 1912— Continued. 

Reflexes: No change from note of eleventh fasting day. 

Chest: No change from note of eleventh fasting day. 

Heart: Measurements of cardiac dullness the same as noted on the 

nineteenth fasting day. The cardiac sounds are rather more distant. 

The aortic second sound is a little louder than the pulmonic sound. 
Pulse: No change from note of eleventh fasting day. 
Abdomen: Very slight gurgling (gas and liquid) with palpation of the 

right hypochondrium. 
Liver : No change from note of twenty-first fasting day. 
Stomach : No change from note of twenty-first day, except that rhythmic 

sounds are heard. 
Kidneys: No change from note of twenty-first fasting day. 
No change noted in glands, lungs, spleen, or genital organs. 
May 8, 1912 (twenty-fifth day of fast): 

General appearance : No change from note of twenty-first fasting day. 
... Mouth: Mucous membrane moist. Good color. Tongue nearly clean. 

Very little odor to breath. 
Reflexes : No change from note of eleventh fasting day. 
Chest : No change from note of eleventh fasting day. 
Heart : Left border of cardiac dullness 6.8 cm. from mid-sternum. Right 

border 0.3 cm. to left of mid-sternal line. Total width 6.5 cm. 

(fifth change noted in measurements of heart). First sounds of the 

heart not distinct. 
Pulse : No change from note of eleventh fasting day. 
Abdomen: Retracted. Dull to percussion except over area of stomach 

tympany. No gurgling on palpation and no tenderness except over 

the pulsating aorta. 
Liver : Lower border of liver dullness unchanged. Total width of dullness 

8.6 cm. (fourth change noted in size of liver). 
Stomach: No change in area of stomach tympany from note of twenty- 
first fasting day. No rhythmic sounds heard. 
Kidneys: No change from note of twenty-first fasting day. 
No change noted in glands, lungs, spleen, or genital organs. 
May 10, 1912 (twenty-seventh day of fast): 

General appearance : No special change from note of twenty-first fasting 

day. 
Mouth: Tongue nearly clean. Odor of breath not specially marked. 
Reflexes : No change from note of eleventh fasting day. 
Chest : No change from note of eleventh fasting day. 
Heart : No change from note of twenty-fifth fasting day. 
Pulse : No change from note of eleventh fasting day. 
Abdomen : No change from note of twenty-fifth fasting day. 
Liver: No change from note of twenty-fifth fasting day. 
Stomach : No change from note of twenty-first fasting day. 
Kidneys: No change from note of twenty-first fasting day. 
No change noted in glands, lungs, spleen, or genital organs. 
May 12, 1912 (twenty-ninth fasting day): 

General appearance: No change from note of twenty-first fasting day. 

Mouth : No change from note of twenty-seventh fasting day. 

Reflexes : No change from note of eleventh fasting day. 

Chest: No change from note of eleventh fasting day. 

Lungs: The percussion note at both apices and just below the clavicle is 

slightly higher pitched. No change noted in the respiratory murmur. 



60 A STUDY OF PROLONGED FASTING. 

May 12, 1912— Continued. 

Heart: No change from note of twenty-fifth fasting day. 

Pulse : No change from note of eleventh fasting day. 

Abdomen: Aside from a slight gurgling of gas and liquid in the right 
hypochondrium with palpation no change noted from twenty-fifth 
fasting day. 

Liver: Total width of liver dullness 7.4 cm. (fifth change noted in size 
of liver). 

Stomach: No change from note of twenty-first fasting day. Rhythmic 
sounds heard. 

Kidneys: No change from note of twenty-first fasting day. 

No change noted in glands, spleen, or genital organs. 
May 14, 1912 (thirty-first day of fast): 

General appearance: The features have a drawn appearance. Subject 
stands erect. Walks with body erect and shows no abnormalities in 
gait but moves about slowly as if fatigued. The skin has a muddy 
yellowish appearance but is soft and moist. The skin is relaxed and 
is easily picked up, as if there were but little subcutaneous fat. The 
muscles are rather small and not firm. There is a slight conjunc- 
tivitis of both eyes. No abnormal pulsation noted in the neck or 
chest, but the pulsation of the abdominal aorta is marked, especially 
when reclining. 

Mouth : Mucous membrane of the lips and cheeks of good color and moist. 
Tongue moist, with a very slight coating on the posterior part. Teeth 
show a slight deposit of discolored tartar at the base. The breath 
has a very slight odor. No enlargement of the tonsils. The pharynx 
shows some injection of the blood vessels and a few blebs on the 
posterior pharyngeal wall. There is a little mucus on the posterior 
naso-pharynx. 

Glands: Cervical, axillary and epitrochlea glands not palpable. A few 
small glands in both groins. 

Reflexes : Pupillary, cremasteric, and plantar reflexes normal. Abdominal, 
patella and Achilles reflex not obtained. No ankle clonus. No 
Rhomberg. 

Chest: Symmetrical. Ribs prominent. Marked sinking in of the supra- 
and infra-clavicular spaces. Good expansion with inspiration. 

Lungs: Percussion of the right lung shows normal resonance to the lower 
border of the sixth rib in the axillary line, to the eleventh rib in the 
mid-axillary line and the eleventh rib in the back. On the left normal 
resonance to the eighth rib in the mid-axillary line and to the eleventh 
interspace in the back. The percussion note is high pitched at 
both apices. Vocal fremitus is slightly increased at both apices, 
more especially on the right. The respiratory murmur is normal 
throughout. 

Heart: The upper border of cardiac dullness is at the fourth rib; the left 
border of cardiac dullness 6.8 cm. from the mid-sternum; the right 
border 0.3 cm. to the left of the mid-sternal line. Total width, 6.5 
cm. The cardiac sounds are distant; the first sound of the heart 
is not clear. The aortic second sound is somewhat sharper than the 
pulmonic second. 

Pulse: The pulses are equal and regular, but small volume. No sclerosis 
of vessels noted. 



PHYSICAL CONDITION OF SUBJECT DURING FAST. 61 

May 14, 1912— Continued. 

Abdomen: Symmetrical. Flat when standing. Much retracted when 
reclining. Marked visible pulsation of the abdominal aorta. The 
abdomen is everywhere flat to percussion except over the area of 
stomach tympany. There is no tenderness except over the abdominal 
aorta. Intestines about the size of a thumb ; can be rolled under the 
finger extending from the caecum to the right hypochondrium and 
from the left hypochondrium down to the brim of the pelvis. There 
is slight gurgling with pressure in the right hypochondrium. 

Liver: The upper border of liver dullness is from the lower border of the 
sixth rib to the costal margin. Total width of liver dullness 6 cm. 
Edge not felt. 

Stomach: Tympany from a point 2 cm. above tip of the ensiform to a 
point 8 cm. above the umbilicus. Total width 9.5 cm. Left border 
of tympany 13 cm. from median line. Rhythmic sounds heard. 

Spleen: The upper border of splenic dullness is at the eighth rib. Area 
of splenic dullness vaguely determined as 6X5 cm. The spleen 
was not felt. 

Kidneys: Right kidney readily palpated. The pole of left kidney is felt. 

Genital organs: Aside from a long prepuce and a slight variococele on the 
left, the penis and testicles are normal. 
May 15, 1912 {two hours after breaking fast) : 

General appearance: Features slightly drawn. Walks erect but delib- 
erately, as if fatigued. No unsteadiness in gait. The tongue is clean 
and there is no odor to the breath. Otherwise no change in the 
physical examination from notes of the thirty-first and last day of fast. 
May 16, 1912 (24 hours after breaking fast) : 

General appearance: Features quite drawn. Walks hesitatingly and a 
little unsteadily. Marked lassitude. Voice weak and faltering. 

Mouth: Mucous membrane of lips and tongue dry. Tongue slightly 
coated posteriorly. 

Abdomen: When subject is standing, abdomen is symmetrical and slightly 
more prominent than on last day of fast. When reclining, the 
abdomen is not so retracted. The right half of the abdomen is 
somewhat more prominent than the left. The right half of the 
abdomen is tympanitic; the left half flat to percussion. Palpation 
of the right half elicits some rumbling of gas and liquid and causes 
slight pain. Rhythmic sounds are heard on the right half. 

Liver: Upper border of dullness at upper border of sixth rib, lower border 
at the costal margin. Total width, 7.5 cm. (1.5 cm. wider than on 
last day of fast). Edge not palpable. 

Stomach: Tympany from a point 2.5 cm. above ensiform to a point 8 cm. 
above the umbilicus. Total width, 10 cm. Left border of tympany 
17 cm. from median line. Rhythmic sounds heard. Otherwise the 
physical examination is the same as on the last fasting day. 
May 18, 1912 (3 days after breaking fast) : 

General appearance : Features appear thin but not drawn. Walks slowly 
but with no unsteadiness. Voice more natural. 

Mouth: Tongue is clean. 

Heart: The left border of cardiac dullness is 7 cm. from the mid-sternal 
line, the right border at the mid-sternal line. Total width 7 cm. 
(0.5 cm. larger than on last fasting day.) Both sounds of heart 
distant but clear and distinct. 



62 A STUDY OF PROLONGED FASTING. 

May 18, 1912— Continued. 

Pulse : Good volume to pulse. 

Abdomen : Slightly more prominent when standing. Abdomen full when 
reclining. Symmetrical. Tympanitic all over, with no difference 
between the two sides. 

Liver: Upper border of liver dullness at fifth rib, lower border at costal 
margin. Total width, 11.5 cm. (the same as at the beginning of the 
fast). Edge not felt. 

Stomach: Tympany from a point 4.5 cm. above the tip of the ensiform 
to a point 5.5 cm. above the umbilicus. Total width, 13.5 cm. The 
left border of tympany extends to a point 16 cm. from the median 
line. Rhythmic sounds heard. Otherwise no change from the last 
fasting day noted in the physical examination. 
May 19, 1912 (4 days after breaking fast) : 

General appearance: Features drawn. Conjunctivae injected (from weep- 
ing). Forehead bathed in cold perspiration. Marked tremor to 
hands. Gait unsteady. Walks as if quite weak physically. (Gen- 
eral condition that of hysteria.) 

Tongue: The tongue is clean. 

Heart: Left border of cardiac dullness 7 cm. from median line. Right 
border 0.5 cm. to right of mid-sternum. Total width, 7.5 cm. 
Sounds clear and more distinct. Aortic and pulmonic second 
sounds of equal intensity. 

Stomach : Tympany from a point 4.5 cm. above the tip of the ensiform to 
2.5 cm. above the umbilicus. Total width, 15.5 cm. Left border 
of tympany 16 cm. from the median line. Rhythmic sounds heard. 
Otherwise no change from the last fasting day noted in the physical 
examination. 
October 19, 1912 (5 months after breaking fast) : 

General appearance: General aspect not that of a well-nourished man. 
Weight 126 lbs., 15 ounces (57.6 kilograms) (nude). Stands erect. 
Gait normal. Skin has a muddy yellowish tinge, but is soft and 
moist. Slight conjunctivitis of both eyes. Small amount of sub- 
cutaneous fat. Muscles moderate in size and rather soft. No pul- 
sation noted in neck, chest, or abdomen. 

Mouth: Mucous membrane of lips and cheeks moist and of good color. 
The tongue has a slight coat, especially on the posterior portion. 
Deposit of discolored tartar on the teeth. Slight odor to breath. 
Pharynx is reddened, the blood vessels injected, and some mucus 
adherent. 

Glands: Cervical, axillary and epitrochlea glands are not palpable. A 
few small glands in both groins. 

Reflexes : Pupils equal and react normally to light and distance. Abdom- 
inal, cremasteric, patella, Achilles, and plantar reflexes normal. 

Chest: Symmetrical, well formed. Some sinking in of the supra- and 
infra-clavicular spaces. Good expansion with inspiration. No bulg- 
ing of the praecordia and apex beat of the heart is not visible or 
palpable. 

Lungs: Percussion of the right lung shows normal resonance to the upper 
border of the fifth rib in the nipple line, to the lower border of the 
fifth rib in the axillary line, and to the eleventh rib in the back. On 
the left normal resonance to the eighth rib in the mid-axillary line 
and to the eleventh interspace in the back. Vocal fremitus is slightly 
increased and expiration slightly prolonged at the right apex, ex- 



PHYSICAL CONDITION OF SUBJECT DURING FAST. 63 

October 19, 1912— Continued. 

tending to the second rib in front and the spine of the scapula behind. 
There were no rales and the lungs were otherwise negative. 

Heart: Left border of cardiac dullness 9 cm. from mid-sternum. Right 
border 2 cm. to right of mid-sternum. Total width, 11 cm. Sounds 
clear. No murmurs. Aortic and pulmonic second sounds of equal 
intensity. 

Pulse : The pulses were equal, regular at rate of 82 per minute. Rhythm 
regular, volume fair. No sclerosis of the vessels noted. Systolic 
blood pressure, 120 mm. Hg., diastolic, 85 mm. (Riva-Rocci instru- 
ment, sitting position.) 

Abdomen : The abdomen is symmetrical, rather prominent when standing 
but flat when reclining. It is soft, tympanitic, but no distension. 
There is no tenderness on palpation. Nothing abnormal felt. 

Liver: The upper border of liver dullness at the upper border of the sixth 
rib. Dullness extends to 2 cm. below costal margin. Total width 
of dullness 11 cm. Edge indistinctly palpable, and with no irregu- 
larities. 

Stomach : Tympany in the median line extended from the tip of the ensi- 
form to a point 3.5 cm. above the umbilicus, a total distance of 11 cm. 
The left border of tympany extended to a point 16 cm. from the 
median line. Faint rhythmic sounds are to be heard with the stetho- 
scope. There is no splashing with palpation. 

Spleen: The upper border of splenic dullness is the eighth rib. Area of 
splenic dullness vaguely determined as 7X5 cm. Spleen not felt. 

Kidneys: Neither kidney palpable. 

Genital organs: Aside from a long prepuce and a slight left variococele, 
the penis and testicles are normal. 

SUMMARY AS TO PHYSICAL CONDITION. 

General appearance: The general appearance of the subject remained 
good throughout the period of observation. A gradual loss of body 
tissue was evident, but the changes were not marked from day to day. 
The most pronounced change was in the abdomen, which became flat 
as soon as he ceased to take food and was distinctly retracted after 
the fifth fasting day. This for the most part appeared to be due to 
the prompt disappearance of gas in the intestines. The actual loss 
in body tissue appeared to be quite evenly distributed over the body, 
but was most noticeable in the tissue of the back of the neck, in the 
sinking in of the supra- and infra-clavicular spaces, and in the promi- 
nence of the ribs. 

The muscles of the extremities, which were but moderately firm at the 
beginning of the fast, appeared to have softened to a slight degree. 
The muscular movements became less active after the seventh fasting 
day, but the impression was that of muscular fatigue rather than 
weakness. The gait was always steady and there was no swaying 
of the body while standing with the feet together and the eyes closed. 
The features frequently appeared drawn after the first week, but this 
was present, as a rule, only during periods of mental depression. 
The tremor of the hands, the weakness of the muscular movements, 
and the changes in the voice noted at the end of the fast and after 
breaking the fast were apparently a part of his hysterical condition. 
The muddy yellowish tinge to the skin did not change throughout 



64 A STUDY OF PROLONGED FASTING. 

The conjunctivitis present at the beginning improved slightly after 
the ninth day of the fast, but was even more marked after the fast 
was broken, probably because of weeping. 

Mouth : The color of the mucous membrane remained good throughout. 
No change was noted in the teeth (the teeth were not brushed through- 
out the period of observation). The slight coating on the tongue 
became more pronounced until the ninth fasting day, when it began 
to disappear slowly. The tongue did not become entirely clean 
until the third day after food was taken. On the third day the odor 
of the breath was offensive, becoming fetid. After the ninth day this 
was less pronounced and gradually decreased until the twenty-third 
day, after which time very little odor was noticed. 

On the ninth fasting day the mucous membrane of the mouth and lips was 
dry. On the eleventh day the lips were desquamating (at the same 
time the seborrhea of the scalp appeared) and this continued until 
the fifteenth day of the fast and did not occur again. The decrease 
of these signs promptly followed the prescribed intake of larger 
quantities of water. No change was noted in the chronic naso- 
pharyngitis. 

Glands : No change whatever was noted in the cervical, axillary, epi- 
trochlea glands, or the glands in the groins. 

Nervous system: During the first week of the fast the mental attitude 
was a cheerful one. He was enthusiastic about the experiment but 
very opinionated. From this time on to the end of the fast he was 
frequently depressed and sometimes irritable. On these days he was 
disinclined to talk, his physical movements were more deliberate, 
and he was more sensitive to any physical discomfort, such as 
pressure of the hands. He attributed this to the depressive effects 
of the rain and cloudy weather, but the impression was that he felt 
actual physical fatigue. On the last day of the fast and for the 
remaining period of observation he exhibited varying mental states 
of depression, irritability, and sullenness, weeping at times. Taking 
the period as a whole there was a gradual increasing depression and 
irritability, which, with the onset of abdominal pain, manifested 
itself in hysterical conditions. There was no outward demonstration 
of any mental improvement as a result of the fast. 

Reflexes: The pupillary, plantar, and cremasteric reflexes were normal 
through the fast. The patella reflex gradually diminished and on 
the fifth day could only be obtained by re-enforcement and on the 
ninth day and during the remainder of the fast could not be obtained. 
The abdominal reflexes were absent on and after the ninth day, and 
the Achilles on and after the eleventh day. Five months after, the 
reflexes were normal. 

Chest : No change was noted in the chest, with the exception of the gradual 
sinking in of the supra- and infra-clavicular spaces and the promi- 
nence of the ribs as the subcutaneous tissues disappeared. 

Lungs: As the loss of tissue in the supra-clavicular space progressed, the 
percussion note of the apices became slightly higher pitched. The 
lower border of resonance of the right lung followed the decrease in 
the size of the liver. No change was noted in the respiratory murmur. 

Heart: There was a gradual decrease in the percussion border of the heart 
noted as follows, the total diminution in size of heart during fast 
being 4.2 cm. : 



c . 5 



FASTING 



PLATE 5 




PHYSICAL CONDITION OF SUBJECT DURING FAST. 



65 





From mid-sternum. 


Total width. 


Left border. 


Right border. 




cm. 
9.5 
8.5 
8.0 
7.5 
7.0 
6.8 
7.0 
8.0 
9.0 


cm. 
1.2 
1.0 
1.0 

0.5 

0.0 
0.3 to left 
. 5 to right 
. 5 to right 

1.5 


cm. 

10.7 
9.5 
9.0 
8.0 
7.0 
6.5 
7.5 
8.5 

10.5 






Seventeenth day of fast 

Nineteenth day of fast 

Twenty-fifth day of fast 

Third dav after fast 


Fourth day after fast 


5 months after fast 





Heart sounds: No change was noted in the character of the heart sounds 
until the eleventh fasting day, when the sounds were less distinct and 
a systolic souffle was heard all over the praecordia. This souffle was 
not heard after the fifteenth day, but the sounds remained more dis- 
tant throughout the fast and after the twenty-fifth day the first 
sound of the heart was not distinct. After the twenty-third day the 
aortic second sound was more distinct than the pulmonic second. 
The systolic souffle appeared at the same period as the seborrhea, 
the dry mouth, and desquamating lips and disappeared when the 
water intake was increased. 

Abdomen: On the first fasting day most of the gas disappeared from the 
intestines. The abdomen became retracted and by the fifth day had 
reached its maximum, the pulsation of the abdominal aorta being 
pronounced. After this time it was everywhere flat to percussion 
except over the area of gastric tympany. After the fifteenth day of 
the fast the contracted large intestines could be palpated along the 
course of the ascending and descending colon. On the twenty-third, 
twenty-ninth, and thirty-first days there was a slight gurgling of gas 
and liquid in the right hypochondrium with pressure. 

Twenty-four hours after breaking the fast the ascending colon was dis- 
tended with gas, but the descending colon was still contracted. On 
the third day after breaking the fast the whole abdomen was tym- 
panitic and symmetrical. 

Liver: A gradual decrease in width of liver dullness as measured in the 
nipple line was noted, as shown in the table. The total diminution 
of the width of liver dullness in the nipple line was 5.5 cm. or 47.8 
per cent of the total width at the beginning of the fast. On the third 
day after taking food, the liver was found to be of the same width 
as at the beginning of the fast. 





Total width. 


First day of fast 


11.5 cm., edge palpable. 
10.5 cm., edge not palpable. 

9.5 cm. 
9.0 cm. 

8.6 cm. 

7.4 cm. 
6.0 cm. 

7.5 cm. 
11.5 cm. 

11.0 cm., edge palpable. 


Third day of fast 


Fifteenth day of fast 


Twenty-first day of fast 


Twenty-fifth day of fast 


Thirty-first day of fast 


Three days after breaking fast . . . 
Five months after breaking fast. . 



Spleen. No change was noted in the spleen during the fast. 



66 A STUDY OF PROLONGED FASTING. 

Kidneys: On the thirteenth day of the fast the right kidney was palpable 
and on the twenty-first the left was palpable and both remained so 
during the rest of the fasting period. Five months later neither 
kidney was palpable. 

Genital organs: No change noted in the genital organs during the fast. 

Observations of the physical condition of Breithaupt, who fasted for 6 days, 
of Cetti, who fasted for 10 days, and of Beauts, who fasted for 14 days, 
failed to show any change in the size or position of the organs. 

PHOTOGRAPHIC STUDY OF SUBJECT. 

The most striking external evidence of prolonged inanition with a 
fasting subject is the degree of emaciation. In order to visualize this 
as much as possible as the fast progressed, an extensive series of photo- 
graphs was taken practically once a week. At these times the calo- 
rimeter laboratory was specially warmed with gas stoves, particularly 
in the part of the room where L. was to pose, and screens were put in 
place for the background. The subject undressed and put on a small 
loin cloth; he was then posed on a low pedestal, which was covered 
with black cloth. In the selection of poses we had the valuable advice 
of Professor W. G. Anderson, of Yale University. 

Probably no routine throughout the whole fast pleased the subject 
more than this series of photographs, as he seemed obsessed with the 
desire to have himself photographed. We were accordingly able to 
obtain a large number of photographs of the subject in a variety of 
poses. Several of those obtained on the first and last days of the fast 
are given in Plates 4 and 5. In the latter part of the fast the subject 
became somewhat less sure of his footing and rested lightly against a 
wooden frame. A rough approximation of the measurements of this 
man may be made by using 640 mm. as the inside distance between 
the wooden uprights. Lack of time prevented our adjusting the 
accurate mirror arrangement of Friedenthal 1 for securing photographs 
that could subsequently be measured. It should be considered, 
however, that the chief reason for taking this series of photographs 
was to visualize the general appearance of emaciation and not to furnish 
material for exact measurements of the loss of tissue. This was sup- 
plied by the accurate measurements made according to the regular 
schedule by Professor W. G. Anderson. (See p. 68.) 

In addition to the anatomical photographs, a great many photo- 
graphs were taken of L. at his own desire, since this seemed to be 
the one thing which would amuse him at any time. Accordingly the 
camera was pointed at him several hundred times throughout the 
course of the fast, although admittedly many of these were false expos- 
ures. A considerable number of photographs were thus obtained which 
show him in his environment, some of which are deemed worthy of 
reproduction (see Plates 1, 2, and 3, pages 11, 19, and 31). 

friedenthal, Med. Klinik, 1909, No. 19, p. 1. 



ANTHROPOMETRIC MEASUREMENTS. 

The importance of careful anthropometric measurements for noting 
the diminution in size of the body as the fast progressed has been recog- 
nized by all writers on fasting. Fortunately Professor William G. 
Anderson, of Yale University, was in Boston during the period when 
this fasting experiment was being made and he kindly offered to make 
a series of anthropometric measurements of the subject. In this he 
was assisted by Dr. W. L. Anderson. These measurements were made 
approximately once a week, Dr. W. L. Anderson making a special 
trip from New Haven to complete the series at the end of the fast. 

Professor Anderson reports that "the measurements of the forearm 
are taken with the hand tightly closed and the wrist slightly flexed. 
The measurements of the upper arm are taken at the largest part after 
the elbow is completely flexed and all flexors and extensors contracted 
to their utmost. In taking the measurements of the calf and thigh, we 
select the largest part after the man has contracted the muscles as 
well as he can while in the standing position." 

The measurements for each week are given in table 1, the total 
decreases in the various girths for the whole fast being given in the last 
column. As would be expected, the largest change in girth was at the 
waist, there being a decrease of 153 mm. The girth of the abdomen 
decreased 119 mm., while a large decrease is shown for both thighs 
and for the chest. Certain of the measurements were not made until 
the second or third examination; the losses are therefore inclosed in 
parentheses to indicate that the series of measurements was not 
complete. The distinct loss in practically all measurements is obvious. 

In the hope of securing some evidence in regard to the muscular 
strength of the subject, Professor Anderson brought with him his 
dynamometers to test the strength. To our great surprise, the subject 
even before the fast began refused to carry out any of these tests, 
stoutly maintaining that he was not an athlete but a professional 
gentleman and that he was not accustomed to doing muscular work of 
any kind. This was wholly in line with his attitude toward other 
muscular- work tests which were contemplated, but which it was neces- 
sary to omit, greatly to our regret. The only evidence that we have 
regarding the muscular strength of the subject is the material obtained 
in the dynamometer tests which were secured every afternoon by Profes- 
sor Langfeld. Even regarding these we are somewhat uncertain as to 
whether the subject exerted his greatest strength in all the tests. The 
pressure which he placed upon the dynamometer was clearly influenced 
by his fear that such pressure might give him a little pain, to which he 
was strongly averse. 

67 



68 



A STUDY OF PROLONGED FASTING. 
Table 1. — Measurements of subject L. 



Measurement. 



April 

11, 
1912. 



April 18, 
5th day 
of fast. 



April 25, 
12th day 
of fast. 



May 2, 
19th day 
of fast. 



May 8, 
25th day 
of fast. 



May 14, 
31st day 
of fast. 



Total 
loss. 



Height: 

Standing 

Sitting 

Girth: 

Neck 

Chest, normal 

full 

empty 

Ninth rib, full 

empty 

Tenth rib 

Abdomen 

Waist 

Hips 

Right biceps, extended . . 

flexed 

Right forearm, extended . 
Left biceps, extended 1 . . . 

flexed 

Left forearm, extended . . 

Right thigh 

Right calf 

Left thigh 

Left calf 

Breadth: 

Shoulders 

Chest, full 

empty 

Hips 

Depth: 

Chest, full 

empty 

Abdomen 



mm. 
1707 



376 
879 
930 

828 
874 
787 

800 
780 

251 
279 
254 
254 
284 
262 
488 
335 
488 
345 



mm. 
1707 



371 
871 
937 
823 

881 
787 
785 
785 
749 

241 
269 
244 
251 
272 
254 
465 
335 
470 
338 



254 
188 



251 
183 



mm. 

1707 

874 

368 
856 
904 
813 
866 
775 
770 
742 
696 

234 
259 
236 
241 
267 
246 
450 
323 
457 
320 

419 
279 
254 
312 

241 
211 
163 



mm. 
1704 

884 

361 
825 
876 
792 
848 
767 
762 
757 
673 
813 
226 
249 
234 
229 
262 
241 
432 
310 
432 
315 

424 
272 
254 
307 

241 
211 
160 



mm. 

1707 

884 

338 
805 
864 
787 
825 
759 
749 
686 
648 
805 
221 
236 
229 
224 
246 
241 
427 
305 
406 
302 

417 
269 
254 

282 

229 
201 
170 



mm. 
1707 

881 

335 
800 
851 
782 
820 
754 
749 
681 
627 
792 
211 
239 
224 
218 
239 
229 
394 
300 
399 
295 

419 
254 
239 
279 

229 
203 
152 



mm. 



(—7) 

41 
79 
79 
46 
54 
33 
(36) 
119 
153 
(21) 
40 
40 
30 
36 
45 
33 
94 
35 
89 
50 

(0) 
(25) 
(15) 
(33) 

25 
(8) 
36 



x The subject was left-handed. 



BODY-WEIGHT. 

To the ordinary individual the most striking index of the severity of 
a prolonged fast is the loss in body-weight, the abstinence from food 
resulting in great emaciation. The fact that even in the short space 
of 24 hours the body-weight changes considerably is not so patent, and 
a consideration of these changes is of interest. If the body- weight is 
determined each hour throughout the day, it will be seen that while 
sudden fluctuations accompany the ingestion of food, the voiding of 
urine, or the passing of feces, there is a general tendency for a regular 
fall in the body-weight from hour to hour amounting, with adults, to 
not far from 40 grams per hour. 1 During the night, the decrease in 
the body-weight is regular, although not quite so rapid. Since such a 
tendency is shown in the course of 24 hours, it would normally be 
expected that it would be more especially evident in the 20 or 30 days 
of a fast and that the body-weight would decrease steadily as the fast 
progressed. 

ROUTINE OF OBSERVATIONS. 

The losses in body-weight have usually been recorded in every 
reported fast, whether scientifically controlled or not. Unfortunately, 
however, the observations vary in value, as the weighings have not 
always been made under constant conditions. At times they even 
show a gain rather than a loss. Comparable results in such observa- 
tions may be secured by the following routine : 

The weights should be taken at approximately the same time each day. 

If the subject is not weighed nude, the clothing worn should be approxi- 
mately of the same weight, and its weight should be deducted from 
the total weight recorded, thus giving the true value for the body- 
weight of the subject. 

The bladder should always be emptied immediately or a short time before 
the weighing. 

The amount of drinking water taken prior to the weighing should be 
constant. 

No water should be taken for some hours before the observation is made. 

The weighings should always be made on the same carefully calibrated 
scales and should be checked by a second observer. 

The environmental temperature and the muscular activity should be 
approximately constant throughout the whole period of the fast. 

As usually fasting subjects are very captious, investigators are ordi- 
narily content to control them only in so far as the collection of the 
excreta and the abstinence from food are concerned, without rigor- 
ously insisting upon their remaining in a quiet, closed room during 

'Benedict and Carpenter, Carnegie Inst. Wash. Pub. No. 126, 1910, p. 113. Benedict and 
Joslin, Carnegie Inst. Wash. Pub. No. 176, 1912, p. 90. 

69 



70 A STUDY OF PROLONGED FASTING. 

the whole period of the fast, with a constancy in the muscular activity. 
In the long fasting experiment with L., however, the routine for weigh- 
ing previously outlined was followed very closely. 

The subject emptied the bladder immediately after leaving the bed 
calorimeter each day about 8 a. m. A respiration experiment of three 
or four 15-minute periods was next made with him. This was usually 
finished about 9 h 30 m a. m. He was then carefully weighed on a 
calibrated platform balance, the weighings and records being made by 
Mr. Carpenter and checked by a second observer. (See plate 2, 
figure D, page 19.) The scales used were the so-called "silk" scales, 
capable of weighing 150 kilograms with a sensitivity of 10 grams with 
a full load. The temperature of the calorimeter room was rarely 
below 20° C, but as the subject was used to a warmer climate he was 
especially sensitive to cold. He was therefore not weighed nude, but 
in a cotton union suit and socks which had been washed in distilled 
water. He also wore over this union suit his heavy woolen underwear. 
The exact weight of this clothing was known and deducted from the 
weight shown on the scales. It was not practicable to make the weigh- 
ing directly after he had emptied the bladder, as it seemed undesirable 
to have him stand so long before the respiration experiment began. 

No water was taken during the night, so that when the subject was 
weighed he had been without water for some 12 hours. Furthermore, 
the amount of water taken during the day was approximately constant 
in quantity, i. e., for the first 10 days of the fast 750 c. c. and for the 
remaining days 900 c. c. During the night the subject had remained 
in the calorimeter chamber under constant temperature conditions, 
and as he usually lay very quietly, the activity was at a minimum. 
While the temperature conditions and muscular activity necessarily 
varied somewhat during the day, they were fairly constant, especially 
as the subject was by nature averse to muscular activity. 

As L. was extremely interested in the records of the body-weight 
from day to day, the change in weight was computed and roughly 
plotted daily in the form of a curve on the blackboard in the calorimeter 
laboratory. Any irregularities in the curve would be instantly detected 
and verifications made if necessary. As a matter of fact, such verifi- 
cation never indicated a discrepancy and we have the fullest confidence 
in this series of weights. 

While the time relations were not theoretically ideal, they were as 
nearly so as was practicable with the large number of observations 
necessary to be made simultaneously upon this man. A sample day's 
computation of the loss of weight is as follows : 

Body-weight in cotton underwear, socks, and heavy woolen underwear, kilos. 

9 h 35 m a.m., April 20, 1912 57.37 

Weight of cotton underwear and heavy woolen underwear. 1 .48 

Naked body-weight, 9 h 35 m a.m., April 20, 1912 55 . 89 

Naked body-weight, 9 h 40 m a.m., April 19, 1912 56.37 

Loss in body-weight, April 19-20, 1912 0.48 



BODY- WEIGHT. 71 

DAILY LOSSES IN BODY-WEIGHT IN FASTING EXPERIMENTS. 

The loss in body-weight in fasts of short duration has been exten- 
sively discussed in a former publication. 1 Since the appearance of this 
book, several other fasts have been reported which were but super- 
ficially mentioned there. We have accordingly gathered together in 
table 2 the records of body-weight obtained in a considerable number of 
fasting experiments continuing 14 days or more. The largest number 
of fasting experiments with any one man has been made with the pro- 
fessional faster, Succi. The scientific aspects of these experiments 
have become world-renowned by means of the classical research of 
Luciani, 2 which was carried out in Florence in 1890 and has never been 
equaled as a careful analytical study of prolonged fasting. In the 
course of this report it will be occasionally necessary to call into ques- 
tion Luciani's conclusions, but the reader is particularly requested 
to consider that since the publication of Luciani's work a quarter of a 
century has passed and that the criticisms raised must be chiefly of the 
technique rather than of the interpretation of the results by the Italian 
master. 

Table 2 includes not only the data for the seven fasts of Succi, but 
also the records of the body-weights secured for three other individuals, 
i. e., Jacques, Beaute, and the fasting woman Schenk. In examining 
these records, it will be seen that in some of the experiments Succi 
had a body-weight some 13 or 14 kilograms greater than in others. 
Several of these observations also show actual gains in weight, as, for 
instance, two records in Succi's fast in Florence, one record in the 
Naples fast, and five records for the fasting man Jacques. Usually 
the protocols for the experiments explain these apparent gains as being 
due to changes in the amount of water consumed or in the time of 
weighing. 

An examination of the losses of weight in these fasts shows that in 
general the larger losses were found in the first days of the fast, although 
on the twelfth day of the Paris fast and the eleventh day of the Milan 
fast, Succi lost more than 1 kilogram of weight. The usual losses in 
the later days of the fast were from 300 to 400 grams in a day. Although 
occasionally records are found of a loss of only 100 grams or less in a 
day, such values are open to suspicion and are generally accounted for 
by errors in weighing or lack of control of conditions. These minimum 
losses are by no means a 'priori evidence that the fast was not genuine 
so far as abstinence from food was concerned, since irregularities in the 
amount of drinking water and particularly in the length of time inter- 
vening between the drinking of water and the weighing, irregularities 
in the voiding of urine as compared with the time of weighing, and 

Benedict, Carnegie Inst., Wash. Pub. No. 77, 1907, p. 301. 

2 Luciani, Fisiologia del digiuno. Florence, 1889. Das Hungern. Translation by M. C. 
Fraenkel. Hamburg and Leipsic, 1890. 



72 



A STUDY OF PROLONGED FASTING. 



changes in the environmental temperature or muscular activity will 
of course increase or decrease the regular loss of material. 

Table 2 also gives the records of the body-weights obtained in the 
fasting experiment with our own subject L. The greatest loss shown 
during the 31 days of the fast was 1.04 kilograms on the first day and 
the smallest loss in weight was 0.11 kilogram on the thirteenth day. 



Table 2. — Losses of body-weight by fasting subjects, with initial weight and weiglii 

on each day of fast. 

(Weight given in kilograms.) 



Day of 

fast. 


Succi. 


Paris, 

1886. 


Milan, 
1886. 


Florence, 
1888. 


London, 
1890. 


Naples, 
1892. 


Rome, 
1893. 


Zurich, 
1896. 


Wt. 


Loss. 


Wt. 


Loss. 


Wt. 


Loss. 


Wt. 


Loss. 


Wt. 


Loss. 


Wt. 


Loss. 


Wt. 


Loss. 


Initial wt. 

1st 

2d .. ,. 

3d 

4th 

5th 

6th 

7th ... . 

8th 

9th 

10th. , . 
11th... . 
12th, 

13th 

14th 










63.30 
62.40 
61.00 
69.80 
59.90 
59.30 
58.65 
58.15 
57.65 
57.25 
56.70 
56.25 
55.60 
55.25 
54.85 
54.60 
54.30 
54.10 
53.65 
53.20 
52.80 
52.60 
62.25 
51.85 
51.45 
51.50 
51.30 
51.25 
51.05 
50.45 












65.10 
64.70 
64.00 
63.30 
62.80 
61.20 
60.90 
60.70 
69.30 
59.10 
59.00 
59.00 
58.90 
58.55 
58.20 
58.00 
57.80 
57.60 
57.50 
57.05 
56.50 


0.40 
.70 
.70 
.50 

1.60 
.30 
.20 

1.40 
.20 
.10 
.00 
.10 
.35 
.35 
.20 
.20 
.20 
.10 
.45 
.55 


71.70 




63.00 
59.40 
59.00 
58.00 
57.40 
57.10 
57.00 
56.90 
56.00 
55.70 
55.30 
54.00 
53.20 


3.60 
.40 

1.00 
.60 
.30 
.10 
.10 
.90 
.30 
.40 

1.30 
.80 


61.30 
59.75 
58.95 
58.20 
57.70 
56.85 
56.30 
56.10 
55.60 
55.40 
54.40 
54.30 
54.00 
53.60 
53.10 
52.85 
52.60 
52.10 
51.35 
51.15 
50.90 
50.90 
50.60 
50.15 
49.70 
49.40 
49.00 
48.70 
48.50 
48.20 


1.55 
.80 
.75 
.50 
.85 
.55 
.20 
.50 
.20 

1.00 
.10 
.30 
.40 
.50 
.25 
.26 
.60 
.75 
.20 
.25 
.00 
.30 
.45 
.45 
.30 
.40 
.30 
.20 
.30 


6.90 
1.40 
1.20 
+ .10 
.60 
.65 
.50 
.60 
.40 
.56 
.45 
.65 
.35 
.40 
.26 
.30 
.20 
.45 
.45 
.40 
.20 
.35 
.40 
.40 
+ .05 
.20 
.05 
.20 
.60 


55.80 
54.90 
53.90 
52.80 
52.10 
51.50 
51.00 
50.40 
50.10 
49.80 
49.70 
49.30 
48.90 
48.70 
48.30 
48.10 
47.55 
47.25 
47.10 
46.80 
46.50 
46.30 
45.90 
45.60 
45.40 
45.20 
44.90 
44.60 
44.30 
44.20 
44.10 
43.80 
43.70 
43.50 
43.20 
43.00 
42.75 
42.60 
42.30 
41.70 


0.90 
1.00 
1.10 
.70 
.60 
.50 
.60 
.30 
.30 
.10 
.40 
.40 
.20 
.40 
.20 
.55 
.30 
.15 
.30 
.30 
.20 
.40 
.30 
.20 
.20 
.30 
.30 
.30 
.10 
.10 
.30 
.10 
.20 
.30 
.20 
.25 
.15 
.30 
.60 


63.60 
61.80 
60.60 
59.80 
59.10 
58.20 
57.50 
57.10 
56.90 
57.20 
56.60 
55.90 
55.50 
55.40 
55.30 
55.25 
54.60 
54.00 
53.50 
53.00 
52.40 


1.80 
1.20 
.80 
.70 
.90 
.70 
.40 
.20 
+ .30 
.70 
.60 
.40 
.10 
.10 
.05 
.65 
.60 
.50 
.50 
.60 






68.50 
68.00 
67.55 
67.25 
66.55 


3.20 1 
.50 
.45 
.30 
.70 


65.70 
65.40 
65.00 
64.55 
64.05 


.85* 
.30 
.40 
.45 
.50 


15th 

16th... 
17th. , 
18th... 
19th... 

20th 

21st , 
22d 


53.00 
52.70 
52.50 
52.20 
51.90 
51.60 
51.20 


.20 s 
.30 
.20 
.30 
.30 
.30 
.40 


63.50 
63.00 
62.75 
62.50 
62.20 
61.90 


.55 2 
.50 
.25 
.25 
.30 
.30 










23d 
24 th... 
25th... , 
26th... 
27th, , 
28th 
29th... 
30th, . 
31st 


50.75 
50.20 
50.10 
50.00 
49.65 
49.50 
49.25 
48.75 


.45" 
.55 
.10 
.10 
.35 
.15 
.25 
.50 


















































































































32d 


























33d 


























34th 


























35th 


























36th 


























37th 


























38th 


























39th 


























40th 





















































1 Loss for 3 days. 



2 Loss for 2 days. 



BODY- WEIGHT. 



73 



Table 2. — Losses of body-weight by fasting subjects, with initial weight and weight 
on each day of fast — Continued. 









(Weight given 


in kilograms.) 












Jacques 


, 1888. 


Beaute, 1907. 


Schenk, 1906. 


L., 1912. 


Day of fast. 


































Weight. 


Loss. 


Weight. ] 


jOSS. 


Weight. 


Loss. 


Weight. 


Loss. 


Initial wt. . . 


62.01 




65.61 




56.3 




60.64 




1st 


60.68 


1.33 


64.57 


1.04 


54.4 


1 


.90 


69.60 


1.04 


2d 


59.74 


.94 


63.72 


.85 


53.6 




.80 


58.68 


.92 


3d 


59.23 


.51 


62.77 


.95 


53.2 




.40 


57.79 


.89 


4th 


59.24 


+ .01 


61.96 


.81 


52.5 




.70 


57.03 


.76 


5th 


58.98 


.26 


61.41 


.55 


51.9 




.60 


56.37 


.66 


6th 


58.35 


.63 


60.83 


.58 


51.2 




.70 


55.89 


.48 




58.55 


+ .20 


60.23 


.60 


50.9 




.30 


55.50 


.39 


8th 


56.68 


1.87 


60.04 


.19 


50.6 




.30 


55.08 


.42 


9th 


56.23 


.45 


59.80 


.24 


50.5 




.10 


54.63 


.45 


10th 


56.23 


.00 


59.11 


.69 


50.2 




.30 


54.13 


.50 


11th 


55.80 


.43 


58.64 


.47 


49.9 




.30 


53.88 


.25 


12th 


55.60 


.20 


58.64 


.00 


49.5 




.40 


63.56 


.32 


13th 


54.67 


.93 


58.37 


.27 


49.1 




.40 


53.45 


.11 


14th 


54.67 


.00 


57.78 


.59 


48.8 




.30 


53.15 


.30 


15th 


55.04 


+ .37 






48.4 




.40 


52.84 


.31 


16th 


54.98 


.06 








48.2 




.20 


52.26 


.58 


17th 


55.06 


+ .08 














51.79 


.47 


18th 


54.81 


.25 
















51.50 


.29 


19th 


53.93 


.88 
















51.11 


.39 


20th 


53.93 


.00 
















50.93 


.18 


21st 


53.82 


.11 
















50.49 


.44 


22d 


53.36 


.46 
















50.13 


.36 


23d 


53.00 


.36 
















49.96 


.17 


24th 


52.74 


.26 
















49.62 


.34 


25th 


52.46 


.28 
















49.33 


.29 


26th 


52.40 


.06 
















49.02 


.31 


27th 


51.89 


.51 
















48.70 


.32 


28th 


52.23 


+ .34 
















48.46 


.24 


29th 


51.86 


.37 
















48.10 


.36 


30th 


51.69 


.17 
















47.69 


.41 


31st 




















47.39 


.30 



Every effort was made to secure uniformity of conditions throughout 
this fast, and probably in no long fast with man have these ideal con- 
ditions been so nearly approached. Yet, even with this care, it will be 
seen that the losses were by no means regular from day to day, although 
the variations are not so great as in the other fasts referred to in table 2. 
Unquestionably the loss in weight in a strictly controlled fast may 
be considered a good general measure of the intensity of metabolic 
processes; yet with such wide fluctuations as are shown for L. it hardly 
seems probable that the body-weight can be looked upon as an accurate 
index of the total tissue change. But attempts have frequently been 
made by investigators to establish a mathematical relationship between 
the daily loss in weight and the length of the fast. Luciani, basing his 
conclusions upon the results of his study with Succi in Florence and 
particularly upon two long experiments with dogs, was confident that 
such a mathematical relationship existed. To study the possibilities 



74 



A STUDY OF PROLONGED FASTING. 



1 

(IIOS 


3 


s 


7 


9 


II 


1? 


IS 


n 


19 


21 


23 


2S 


27 


29 


3f 33 35 37 


3< 






\ 














)AY i OF 


Fast 


















670 
8U 




\ 


\ 




































64i 
HA 


\ 


v 








































1 
















\ 


ZURICH 




















\ 


\ 
















v 1896 




















\ 


\ 


\ 






































IV 

\ 1 


\ 


\ 


\ 




































1 


\ 




V 








































\\ 


















































































V 




\ 










































\ 












ROMI 


















WO 






\ 
\ 


\ \ 


V 


v^ 








\ 


89 


3 




















\ 






\ 





\ 


1, 






























\ 








\ 




\ 


























\ 








\ 




s 


L 1 


89 


2 




















'10 




\ 










\ 
\ 

\ 




\\ 
V 

1 


\ 


























V 












^ 


























m 




\ 


\ 
















\ 
\ 


























V 












\ 






'.FLORENCE 










sin 






V 


\ 


















\_ 


IIS 


88 














k 






\\ 




















\ 


/ 












M 
























V. 
























\ 


N 
















1 


\ 

\ 
\ 




w 


kRI 


) 
































1 














411 












\ 




\ 






M 


IL 

1 


J8 


\ 
6^ 


\ 












470 










































4U 










































mti 








































































LONDON 






440 
































1890 





































































































































Fig. 1. — Body -weight curves for fasting experiments 
with Succi. 



BODY- WEIGHT. 



75 









3 5 


7 


9 


ii 13 is n 


19 2 


23 25 27 29 31 






\ILOS 

610 
















D. 


kYS 


OF 1 


"AST 
















! 






\J 




































60.0 
















































































590 
















































































5fl0 
















































































570 




\ 


S 








































\ 


































560 








\ 








































^ 


\\ 






























JV>0 










\ 






































































MO 
















































































•no 
















































































s?n 
















































































51.0 
























*\ 
























































son 
















































































440 
















































































480 
















































































47.0 












































































\ 



Fig. 2. — Body-weight curve for Levanzin. 



76 A STUDY OF PROLONGED FASTING. 

of this relationship, curves have been plotted showing graphically the 
changes in weight during each of Succi's fasts. (See figure 1.) A 
similar curve has been plotted showing the records of body-weights 
during the fasting experiment with L. (See figure 2.) 

If we analyze the components which make up the loss in weight 
of the body, we find it is due not only to the loss of body tissue which is 
oxidized to supply material for the maintenance of the body activity, 
but to the loss of preformed water, i.e., the water existing in the tissues 
oxidized. According to observations in some of the earlier fasting 
experiments in Middletown, Connecticut, this preformed water, which 
varies widely in amount, appears to be rapidly discharged in the first 
days of fasting. We would consequently expect to find that the curves 
for a fasting experiment would indicate a rapid fall in weight at the 
beginning of the fast, the percentage loss becoming gradually smaller, 
until the body-weight curve tends to become a straight line. If the 
curves for Succi and Levanzin are examined, this tendency will be seen. 
On the other hand, while all the curves have the same general trend, a 
careful mathematical analysis shows no regularity that would justify 
the use of a mathematical expression by means of which losses of weight 
may be predicted during prolonged fasts. When one considers that 
only the Florence fast was controlled by Luciani, and that the others 
were made in different years, at different seasons, and in different 
countries, it will be seen that but little can be expected from a com- 
parison of these curves. 

Nevertheless, the semblance of mathematical regularity shown in 
the records of body-weight obtained in Succi's Florence fast and in the 
experiments with dogs led Luciani to seek the aid of his associate, 
Bufalini, who computed that the body-weight curves, especially those 
obtained in experiments on dogs, (see P and P' on figure 3) had a ten- 
dency to represent an equilateral hyperbola. Reasoning from the 
equilateral hyperbola equation obtained with dogs, Luciani computed 
the probable curve for Succi's weights during the Florence fast and 
found that the loss in weight was very much less than he would have 
expected. He interpreted this as being due to the fact that Succi 
drank much larger amounts of water than did the dogs and that water 
apparently acted as a nutrient, thus sparing the tissues. 

Since a reasonable regularity was also shown in the course of the 
curve obtained for L., a probable curve for this subject was developed 
by Mr. E. H. Lange, physicist of the Nutrition Laboratory. (See 
curve in light line in figure 2.) Using W to represent the weight in 
kilograms and T the time in days, the weight for any given day is 
found by the formula: 

TF=3.20 (10)- 0143r -0.3247 7 +57.43 



BODY-WEIGHT. 



77 



A similar equation has been worked out for Succi's London fast, as 
follows : 

TT=4.98 (10)- 00692r -0.2227 7 +50.75 

(See curve in light line in fig. 1.) Obviously such a complicated curve 
can not in any wise be considered a simple mathematical relationship. 



5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 SO 85 90 95 100 105 110 115 120 



PRO 


















































K ft s. 
?7 






















DA 


fS 


f FA 


ST 






















26X1 


\\ 
















































25.0 


v 


\HA 


M 

1 FAE 


T 










































?4.0 


\ 


v 














































?30 


HAW 
1" 


A 

FAST 














































22.0 


















































?1.0 










V 








































20.0 


\a 


WROR 
»3 


3W 




10. 


I 






































19.0 










Kgs. ~* 
1 

0J> 




N 


































iao 


\a 


WROR 
V 4 


DW 




i *, n 






l 
































170 










7.0 






^ 


\ 






























16.0 




AWRO 
S«2 


ROW 




«!<5 — 










v. 




























i<sn 










•v «S.fl 








































1 




met 

\p 


Nl 


4*0 






































13j0 


\ 








3 





















V 


^A 






















Vlu 

V 


JANI 
















































































































































ex> 


















































6.0 














\ 




































7.0 














\ 





































Fia. 3. — Body-weight curves for prolonged fasting experiments with dogs. 

Ideal conditions for studying the loss in weight during fasts would 
be those in which the subject had a constant amount of drinking water, 
emptied the bladder completely at a definite period each day, remained 
in an absolutely constant environment with a constant temperature, 



78 A STUDY OF PROLONGED FASTING. 

and the metabolism pursued a course entirely unaffected by extraneous 
conditions. Such constant conditions are impracticable with human 
beings, but are more easily obtained with animals. Several remarkably 
long fasting experiments have been made with dogs, but the records are 
for the most part not easily accessible. Curves showing the records of 
body-weight in these animal experiments are given in fig. 3. The body- 
weights obtained by Luciani on two dogs, one in Siena and another 
in Florence, are represented by the curves designated as P and P\ 
These dogs were catheterized each day, were given exactly 150 c. c. 
of water daily, and were kept in a room free from disturbance of any 
kind and in a temperature of approximately 12° to 13° C. It will 
be seen that the loss in weight follows a fairly regular course, save on 
the last few days of each fast, none of the irregularities shown in the 
body-weights of Succi and our subject L. being apparent. 

Luciani's two experiments with dogs continued 34 and 43 days 
respectively, but still longer experiments have been made by P. B. 
Hawk, in which a dog fasted in two experiments of 117 and 104 days 
respectively. The complete results of these experiments have not yet 
been published, 1 but the investigator has kindly given me the privilege 
of using certain of the data in this connection. The body-weights 
obtained during these fasts are also shown in figure 3. The amount 
of drinking water given to the dog was constant, but the animal was 
not kept in a chamber with even temperature and the other conditions 
were not so well controlled as in Luciani's experiments, as the purposes 
of the fasts were entirely different. Aside from slight fluctuations, 
however, the curves follow a reasonably constant course. These 
curves are of particular value owing to the extraordinary length of the 
fasting periods. 

A series of observations made by Awrorow on dogs with complete 
fasting is of even more interest in studying this specific problem. 
The dogs were confined in the Pashutin respiration chamber in the 
Imperial Medical Academy in St. Petersburg, receiving neither food 
nor water. They were catheterized daily, weighed at a regular hour, 
and spent 22 or more hours out of the 24 hours in the quiet and isolation 
of the respiration chamber, during which the carbon-dioxide production 
was carefully measured. The period of fasting with the four dogs 
continued for 16, 44, 60, and 66 days respectively. Under these con- 
ditions one would expect a most regular progress in the metabolism, 
with constant loss of water and organic material, and changes in the 
body-weight. That this is true to a marked extent is shown from an 
examination of the three curves for the dogs which fasted the longest, 
i. e., 44, 60, and 66 days. The striking regularity of these curves bears 
out completely Professor Luciani's view that if such experimental con- 

'Howe and Hawk, Am. Journ. Physiol., 1912, 30, p. 174; Howe, Mattill, and Hawk, Journ. 
Biol. Chem., 1912, 11, p. 103. 



BODY-WEIGHT. 79 

ditions can be secured in a fasting experiment, the curve will be extra- 
ordinarily regular. 1 

As an effort was made to secure constant conditions in the experiment 
with our fasting subject, it was hoped that a curve approximating the 
regularity of the curve for Awrorow's dogs could be obtained, but an 
examination of the plotted values for the daily body-weights shows 
that this was far from being the case. It would be practically impos- 
sible to carry out a lengthy fasting experiment with a man with envi- 
ronmental conditions as constant as were those of the dogs used by 
Awrorow. We know also that the loss of material varies greatly both 
in amount and in kind with the progress of the fast. Thus there is 
always greater metabolism, greater activity, and a greater disintegra- 
tion of material in the first few days of fasting. As the fast progresses, 
the effect becomes more or less noticeable, and the subject becomes 
disinclined to active muscular work, thus naturally conserving his 
energy. Furthermore, human subjects, by covering themselves with 
extra clothing and preferring warm rooms, attempt to conserve their 
calorific output. 

The character of the katabolism may also vary greatly, particularly 
during the first few days of the fast, so that there is unquestionably 
a rapid depletion of the glycogen storage in the body in this period. 
The course of the curve would therefore vary according to whether the 
subject of the experiment was well nourished, poorly nourished, or 
obese. The character of the foregoing diet may likewise play a role in 
this connection. It will be seen later, however, when a study is made of 
the gaseous metabolism during the fast, that the possibility for analyz- 
ing the daily losses in a fast such as that carried out by L. are much 
greater than in any fasting experiment thus far observed with man, and 
that no mathematical relationship between the length of time of a 
human fast and the loss in body-weight can be expected. 

J The data for these curves were secured from the large work on fasting dogs written by Professor 
Awrorow. Curves showing the percentage loss of body-weight for two of these dogs are shown on 
two lecture charts kindly given me by Professor Awrorow during my 1907 visit to St. Petersburg, 
and reproduced in figures 45 and 46 on pages 356 and 357 of this report. 

While this report on the long fasting experiment made in the Nutrition Laboratory is designed 
primarily to deal only with the influence of prolonged fasting on human metabolism, it will not 
be out of place here to emphasize the fact that a colossal amount of research on fasting animals 
has been accumulating in the laboratories in St. Petersburg for a number of years, chiefly under 
the direction of Albitsky. A considerable portion of this material was incorporated in the second 
volume of Pashutin's experimental pathology, of which 840 pages are devoted to the discussion 
of fasting. This material has been considered so important to workers in metabolism that it has 
been translated in the Nutrition Laboratory and typewritten copies of the translation have been 
deposited in the library of the Office of the Surgeon-General of the Army, in Washington, D. C, 
the John Crerar Library in Chicago, Illinois, and the New York Public Library in New York City, 
the fourth copy being retained by the Nutrition Laboratory. 

In addition to this material in Pashutin's book, much of which is recorded for the first time, 
there are a number of large monographs published by Awrorow, Kartaschefsky, Albitsky, and 
Likhatcheff , which deal with the abstract problems of metabolism and which have been translated 
in whole or in part for this laboratory. It is greatly to be regretted that this collection of data, 
which far exceeds both in quality and amount the total accumulation of earlier investigators on 
fasting animals, should be so inaccessible to American, English, and Continental readers. It is 
certainly true that no one working on the gaseous metabolism of animals can at the present day 
afford to overlook this wonderful collection of Russian material. 



80 



A STUDY OF PROLONGED FASTING. 



TOTAL LOSS IN BODY-WEIGHT. 

An examination of the data in table 2 for the subjects other than L. 
show apparent discrepancies in the initial weights and in the loss of 
weight on the first day. The exact length of the fasting period and 
the weight on the last day are also frequently doubtful. The Florence 
weights were all taken from the plates at the end of Luciani's report 
"Fisiologia del digiuno." It is obviously important to note whether 
the initial weight was taken immediately after the meal or before 
the meal, or what was the condition of the alimentary tract. With 
our subject L. we believed it to be necessary to obtain an accurate 
weight at the beginning of the fast ; consequently the initial weight was 
taken approximately 12 hours after the last meal, several hours after 
drinking water, and a definite time after urinating. Such precautions 
were not taken, we believe, with any of the other subjects given in 
table 2, with the possible exception of Cathcart's subject, Beaute\ 

The total losses in the various fasts, particularly when computed as 
percentages of the initial weights, have certain features that are not 
without interest. For comparison we give in table 3 the total loss and 
the percentage loss for each subject at the end of 14, 16, 20, 29, 30, 31, 
and 40 days respectively (using the data recorded in table 2). Obvi- 
ously a comparison can be made for all of the subjects for only 14 days, 
with all the subjects but one for 16 days, with all the subjects but two 
for 20 days, and finally with but one subject for 40 days. At the end 
of 14 days the average percentage loss was 12.6 per cent. The lowest 
loss was with Succi, in the Rome fast, of 10.6 per cent; and the highest 
loss was 15.7 per cent with the same subject, in the Paris fast. For 





Table 3. — Summary of losses of body-weight by fasting subjects. 








Subject. 


14 days. 


16 days. 


20 days. 


29 days. 


30 days. 


31 days. 


40 days. 


Kilos. 


Per 
cent. 


Kilos. 


Per 

cent. 


Kilos. 


Per 
cent. 


Kilos. 


Per 
cent. 


Kilos. 


Per 

cent. 


Kilos. 


Per 

cent. 


Kilos. 


Per 
cent. 


Levanzin . . 
Succi: 

Paris. . . . 

Milan . . . 

Florence . 

London. . 

Naples . . 

Rome . . . 

Zurich.. . 
Jacques. . . . 
V.BeautS.. 
Schenk .... 

Average . 


7.49 

9.90 
7.70 
8.45 
7.10 
8.20 
6.90 
7.95 
7.34 
7.83 
7.50 


12.4 

15.7 
12.6 
13.3 
12.7 
12.9 
10.6 
11.1 
11.8 
11.9 
13.3 


8.38 

10.30 
8.45 
9.00 
7.70 
8.35 
7.30 
8.70 
7.03 


13.8 

16.3 
13.8 
14.2 
13.8 
13.1 
11.2 
12.1 
11.3 


9.71 

11.40 

10.15 

10.50 

9.00 

10.60 

8.60 

9.80 

8.08 


16.0 

18.1 
16.6 
16.6 
16.1 
16.7 
13.2 
13.7 
13.0 


12.54 

13.75 
12.80 
12.85 
11.50 


20.7 

21.8 
20.9 
20.3 
20.6 


12.95 

14.25 
13.10 


21.4 

22.6 
21.4 


13.25 


21.9 


























11.60 


20.8 


11.70 


21.0 


14.10 


25.3 


































10.15 


16.4 


10.32 


16.6 


















8.1 


14.4 












































12.6 




13.4 




15.6 




20.1 




20 6 




21.5 




25.3 









BODY-WEIGHT. 81 

16 days the average percentage loss was 13.4 per cent, the lowest again 
appearing with Succi in the Rome fast of 11.2 per cent and the highest 
16.3 per cent in the Paris fast. For 20 days the average percentage 
loss was 15.6 per cent, the lowest being with Jacques of 13 per cent, 
and the highest 18.1 per cent with Succi in the Paris fast. For 29 
days the average loss was 20.1 per cent, the lowest again being with 
Jacques of 16.4 per cent and the highest 21.8 per cent in the Paris fast 
of Succi. For 30 days the average value was 20.6 per cent, the lowest 
being with Jacques, of 16.6 per cent, and the highest 22.6 per cent in the 
Paris fast of Succi. For 31 days only two experiments were com- 
parable, the percentage loss in both of these being about 21 per cent, 
while in the 40-day experiment the percentage loss was 25.3 per cent. 
Of special interest is the fact that, aside from the fast with Jacques, in 
which the weights of the drinking water and urine were perhaps less 
trustworthy than in the other fasts, it can be said that at the end of 30 
days of fasting, 21.5 per cent of the initial body- weight was lost. This 
is strikingly regular in the fasts of both Succi and Levanzin. The 
maximum loss of weight recorded in any controlled fasting experiment 
with man was in the London fast with Succi, when at the end of 40 
days a loss of 25.3 per cent of the initial body-weight was shown. 

In contrast with these values found with man are the losses found 
with animals, when the degree of emaciation has been carried to an 
extreme and, indeed, in some instances to the point of death. In 
Hawk's first fasting experiment, in which the dog fasted 117 days, 
62.9 per cent of the initial body-weight was lost. The dog recovered, 
was fed, and later underwent a second fast of 104 days, in which he 
lost 52.5 per cent of the initial body-weight and then suddenly died. 
Hawk's dog lived in the laboratory and was given a definite amount of 
water, but Awrorow's dogs received neither food nor water, and the 
fasting was carried to the point of death. With dog No. 2 the fast 
lasted 44 days, with a loss of 55 per cent of the initial body-weight. 
With dog No. 3 the fast continued 60 days, with a loss of 61.6 per cent, 
while with dog No. 4 the fast was 66 days in length, with a loss of 62.0 
per cent of the initial body- weight. Still other values were obtained 
by Luciani with the two dogs which fasted under special experimental 
conditions. As shown in his curve, 1 the dog P lost 43.5 per cent of his 
initial body-weight in a 43-day fast, while the dog P' lost 45.5 per cent 
in a 34-day fast. 

Incidentally it should be mentioned that Gayer, in his 30-day fast in 
New York in 1912, was said to have lost 17.4 per cent of his initial 
weight of 210 pounds (95.3 kilograms), while Penny, in his self-con- 
trolled fast of 30 days, lost 19 per cent of his initial body- weight of 
137.5 pounds (62.4 kilograms). While both these values are somewhat 

luciani, Das Hungern, Hamburg and Leipsic, 1890, plate n. See also this publication, fig. 
3, p. 77. 



82 A STUDY OF PROLONGED FASTING. 

less than the average loss found with L. and Succi, they are sufficiently 
close to imply that in all probability no measurable amounts of food 
were taken during these two uncontrolled fasts. 

ANALYSIS OF LOSSES IN BODY-WEIGHT. 

An analysis of the factors influencing the body-weight shows that 
there may be a retention of water in the body due to the drinking of 
more water than is excreted in the urine; a loss due to feces; a regular 
loss due to the oxidation of organized material, the carbon burning 
to carbon dioxide and the hydrogen to water; and a further loss of solids 
in the urine. The amount of organized material oxidized in the body 
will be influenced in large part by the muscular activity of the subject, 
and if the activity is constant, the loss due to oxidation will progress in 
a reasonably regular manner. 

Considering the body as a living organism, therefore, we see that in 
a fasting experiment the intake consists of drinking water and oxygen 
from the air. The output consists of water-vapor and carbon dioxide 
given off from the lungs and skin and the urine and feces excreted. 
In this particular fast, however, the subject did not defecate during the 
experiment. 

The water vaporized from the lungs and the skin and given off in the 
urine undoubtedly contains a large amount of preformed water which 
was taken with the water drunk each day. It also contains water 
which has been stored in the body and is given off as a result of the 
breaking down of the protein, i. e., muscular tissue. There is likewise 
a small amount of water due to the combustion of the organic hydrogen 
of the body with the oxygen taken from the air. Without estimations 
of the carbon-dioxide excretion, there are at present no known means of 
satisfactorily computing these separate factors in the measurement of 
the water output. When it is possible to have the subject live the 
entire time inside the respiration chamber, as was done in the experi- 
ments at Wesleyan University, 1 the complete income and outgo may 
be determined, including the income of oxygen and water and the out- 
put of carbon dioxide, water- vapor, water in urine, solids in urine, and 
an analysis of the solids. An approximate apportionment may then 
be made of the water leaving the body as oxidized organic hydrogen 
and as preformed water in the body. This has already been done 
for the 7-day experiment reported in the earlier publication. 2 From 
the computed amounts of carbon dioxide excreted and the probable 
organic hydrogen oxidized, a similar apportionment of the water loss 
has been made for this experiment (see page 407 of this report). 

Inasmuch as the body consists in large part of water — some 60 per 
cent or more — it will be seen that there may be an addition to or loss 

Benedict, Carnegie Inst. Wash. Pub. No. 77, 1907. 
"Benedict, Carnegie Inst. Wash. Pub. No. 77, 1907, p. 469. 



BODY-WEIGHT. 83 

from the storage of water in the body, as, for instance, 200 grams, 
without materially affecting the total percentage of water. It is easy 
to see, therefore, that the changes in weight noted from day to day with 
a fasting subject have only an indirect and passing influence. 

INSENSIBLE PERSPIRATION. 

In the long fasting experiment with L., the subject was not kept 
inside the respiration chamber for the entire time of the fast, so that the 
complete output of water-vapor was not determined. On the other 
hand, a study of the so-called "insensible perspiration, " which has been 
of great interest ever since the days of Sanctorius, shows some facts of 
value. 

The body loses in weight regularly as a result of the elimination of 
carbon dioxide and water-vapor. It loses weight spasmodically by 
the passing of urine and it gains in weight spasmodically by the drink- 
ing of water. By correcting for the amount of water taken and the 
weight of urine passed, the degree of insensible loss, or the "insensible 
perspiration," may be accurately calculated. This has been done in 
table 4, which gives for each day of the fast the loss of body-weight 
in grams, the weight of the urine passed, the weight of the drinking 
water taken, and the insensible perspiration. The excretion of urine 
was always less than the amount of the drinking water with one excep- 
tion, that of April 29-30. The insensible perspiration is therefore 
readily obtained by finding the difference between the amount of water 
taken and the weight of urine excreted and adding it to the observed 
loss in body-weight. 

A fact of special interest in this connection is that while the losses 
in body- weight fluctuate considerably, the losses as shown by the insen- 
sible perspiration are reasonably regular, the lowest being 371 grams 
on May S-A; after the first 10 days, the highest value was 691 grams on 
April 25-26. Theoretically this insensible perspiration should give us 
a reasonable clue to the progress of the fast and should be an index of 
the loss of water and carbon dioxide. On the other hand, while the loss 
of preformed water is a real quantitative loss, the carbon dioxide and 
water of oxidation are not, as they are in large part made up of oxygen 
which is taken from the air. 

In the later part of the fast it will be seen, from table 4, that this man 
had on the average an insensible perspiration of not far from 20 grams 
per hour. The insensible perspiration of the subjects in the fasts 
described in the earlier publication, particularly S. A. B., was inad- 
vertently not reported. Subsequently, Benedict and Carpenter, 1 in 
discussing the metabolism of healthy men, computed the insensible 
perspiration of all of the fasting subjects. From their figures it will 
be seen that the fasting subject S. A. B., who spent 24 hours of each day 

1 Benedict and Carpenter, Carnegie Inst. Wash. Pub. No. 126, 1910, p. 114. 



84 



A STUDY OF PROLONGED FASTING. 



inside the respiration chamber, had an insensible perspiration of not 
far from 25 to 27 grams per hour. When it is considered that this 
represents the first 5 to 7 days of fasting, it will be seen that the results 
obtained for L. are quite in accordance with those secured with the 
earlier subject, in that they indicate a distinct tendency for the insen- 
sible perspiration to decrease as the fast progressed. It should be 

Table 4. — Insensible perspiration during fasting experiment with subject L. 



Date. 


Day of 
fast. 


Loss of 
body-weight. 

A 


Urine. 
B 


Drinking 
water. 

C 


Insensible perspiration. 


Per 24 hours 

a+(c-b). 

D 


Per hour. 
E 


1912. 

Apr. 14-15 

15-16 

16-17 

17-18 

18-19 

19-20 

20-21 

21-22 

22-23 

23-24 

24-25 

25-26 

26-27 

27-28 

28-29 

29-30,.. 
Apr. 30-May 1 . . 
May 1-2 

2-3 

3-1 

4-5 

5-6 

6-7 

7-8 

8-9 

9-10 

10-11 

11-12 

12-13 

13-14 

14-15 


1st 

2d ,, 

3d 

4th 

6th. .. . 

6th. . . . 

7th. . . . 

8th.... 

9th 

10th.... 
11th.... 
12th.... 
13th.... 
14th.... 
15th.... 
16th.... 
17th.... 
18th.... 
19th.... 

20th 

21st . . . . 

22d, 

23d. 

24th 

25th.... 

26th 

27th 

28th 

29th 

30th.... 
31st 


grams. 
1040 
920 
890 
760 
660 
480 
390 
420 
450 
500 
250 
320 
110 
300 
310 
580 
470 
290 
390 
180 
440 
360 
170 
340 
290 
310 
320 
240 
360 
410 
300 


grams. 
674 
482 
681 
731 
683 
624 
537 
601 
622 
678 
677 
529 
674 
660 
768 
902 
861 
669 
740 
709 
717 
795 
566 
760 
722 
737 
663 
663 
706 
780 
675 


grams. 
720 
750 
750 
750 
750 
750 
750 
750 
750 
760 
900 
900 
900 
900 
900 
900 
900 
900 
900 
900 
900 
900 
900 
900 
900 
900 
900 
900 
900 
900 
900 


grams. 
1086 
1188 
1059 
779 
727 
606 
603 
569 
578 
672 
573 
691 
436 
540 
442 
578 
509 
521 
550 
371 
623 
465 
504 
480 
468 
473 
657 
477 
554 
530 
625 


grams. 
45 
60 
44 
32 
30 
25 
25 
24 
24 
28 
24 
29 
18 
23 
18 
24 
21 
22 
23 
15 
26 
19 
21 
20 
19 
20 
23 
20 
23 
22 
26 



borne in mind, however, in making any comparisons between the results 
obtained in these two fasting experiments, that while the subject 
S. A. B. remained in a respiration chamber the whole period of the fast 
and consequently had an approximately constant temperature envir- 
onment and minimum muscular activity, the subject L. was not in a 
respiration chamber, but was for certain days partially naked for some 
time while being photographed, measured, or clinically examined, 
occasionally had a bath, and several times went out for a carriage 



BODY-WEIGHT. 85 

drive. His temperature environment and muscular activity were 
therefore more variable than those of the subject of the earlier fasting 
experiment. 

An exact explanation of the variations in the insensible perspiration 
from day to day is difficult, particularly for those of May 3^1 and 4-5, 
when the lowest value of the fast was found on the day before a very 
much higher value was found. It is always possible that there may 
have been an error in the weighing, but on the other hand these weigh- 
ings were very carefully made and recorded. Furthermore, an attempt 
to explain the variations on account of a difference in the activity is 
somewhat difficult, since the subject had a drive on May 3-4 and a bath 
and drive on May 4-5, and on other days when he was given a carriage 
ride the insensible perspiration was much greater than on May 3-4. 
It is obvious, therefore, that these figures should not be considered 
individually, but only as a general picture, this showing that the insen- 
sible perspiration had a tendency to decrease as the fast progressed. 
The increased value for May 14-15 may without doubt be explained 
by the fact that this being the last day of the fast, there was much 
greater excitement and muscular activity on the part of the subject. 
On this day he talked vigorously to a group of medical men for some 
40 or 50 minutes; on this day, also, he was nude for a time while photo- 
graphs were being taken and during a series of physical measurements. 
Considering the values generally, however, it will be seen that the insen- 
sible perspiration is a far more scientific basis for estimating the loss of 
body-substance during a fast than is the mere record of body-weight, 
which considers neither fluctuations in drinking water nor the volume 
of urine passed. 

DRINKING WATER. 

The intake of a fasting man is confined to water and oxygen of the 
air. Of these the water may be readily measured. Such measure- 
ments are of great importance in intelligently interpreting the losses 
in weight from day to day. Accordingly special care was taken to 
insure accurate records of the water consumed. 

The selection of the kind and amount of drinking water for use in a 
long fast is by no means simple. On the one hand, there is the belief 
that distilled water is dangerous in that it washes out the salts from the 
body, while on the other there is the fact that in many fasts the subjects 
took either ordinary tap water or, as in Succi's fasts, various alkaline 
or spring waters containing large amounts of salts, sometimes of a 
distinctly purgative character. It has been believed by some that 
these mineral waters have an actual nutritive value due to the salts 
contained in them, if not to the organic matters. Furthermore, the sup- 
position is reasonable that the salts interfere seriously with the mineral 
metabolism, and it is obviously impossible in a metabolism experiment 
to make an intelligent study of the output of sulphur, phosphorus, or 



86 A STUDY OF PROLONGED FASTING. 

chlorine in the urine if at the same time the subject is taking a large 
amount of water containing sulphates, phosphates, or chlorides. 

In the fasting experiments at Wesleyan University, one of the 
subjects, S. A. B., preferred distilled water. 1 Similarly Penny 2 records 
that he used only distilled water during his fast. Our subject, L., 
himself suggested that he be given distilled water during the fast, as 
otherwise it might be said that tap water was either not pure or con- 
tained mineral or organic matters which would contribute to his sus- 
tenance. Although an experiment in which the man used distilled 
water only was somewhat unusual, the desirability of being able to 
study the mineral metabolism without the conflicting factor of the 
ingestion of salts was, of course, apparent and arrangements were 
therefore made for supplying L. with distilled water throughout the fast. 

Dr. E. P. Cathcart was at this time a Research Associate of the 
Nutrition Laboratory and advised that L. be given a constant amount 
of drinking water each day, since in his observations on Beauts he had 
experienced considerable difficulty with the volumes of the urine. 
Our subject was first given 1,000 c.c. of distilled water, but was able 
to take but 720 c.c. on the first day. L. then suggested that he be 
given only 750 c.c. This amount was continued for a number of days, 
when it was increased to 900 c.c, at which volume it continued through- 
out the remaining days of the fast. 

The amount of water taken each day by the subject is given in 
table 4. Since the body excretes so large an amount of water, it is 
perhaps somewhat unfortunate that the volume taken by the subject 
was not constant for the whole period of the fast, although it is much 
more nearly uniform than in any long fast heretofore reported. In any 
discussion of the body-weight or the volume of urine, it is obviously 
necessary to consider these fluctuations in the intake of water. 

The subject was very inconsistent in his comments regarding the 
water. On some days he said it was very good, but on other days 
considered it to be very bad, although exactly the same amount was 
given him and from the same glass carboy. On some days, also, he 
found the amount given him was not enough and again not infrequently 
complained that he was given too much water. He recognized the 
importance, however, of maintaining the volume of urine so that a large 
number of analyses could be made. 

The daily allotment of distilled water was supplied to the subject in 
a bottle and from this he poured out the amount desired. Early in the 
fast he found that it was desirable to drink as large a portion of water 
as possible during the first part of the day, so that it would not be neces- 
sary for him to urinate during the night. The records show that he rarely 
urinated during the night. Furthermore, as in the later part of the 

Benedict, Carnegie Inst. Wash. Pub. No. 77, 1907, pp. 136 and 140. 
'Penny, British Med. Journal, 1909, p. 1414. 



BODY-WEIGHT. 87 

fast he divided the urine into day and night periods, it provided a 
particularly satisfactory method for studying the constituents of the 
urine separately for these periods. Usually the last of the water was 
taken a short time before the subject entered the bed calorimeter for 
the night experiment. 

At the end of the first 10 days of the fast, during which L. had taken 
but 750 c.c. of water daily, the attending physician, Dr. H. W. Goodall, 
expressed the opinion that there was a distinct physiological need of 
water in the body. The lips of the subject were parched, his skin was 
dry, and dandruff appeared. At Dr. Goodall's suggestion, L. was 
prevailed upon to increase the amount of drinking water to 900 c.c. 
daily. Two days later he reported to Dr. Goodall that for the first 
time he felt thirst. Unfortunately some of the statements of the 
subject were so inconsistent at this time that it is difficult to say 
whether or not there was a physiological need for water which was not 
felt by the subject but which was observed by the physician. 

Aside from the objective indications noted by Dr. Goodall, the need 
for water in the body may be inferred, though not scientifically proved, 
by the figures given in table 4, the difference between the water taken 
and the urine excreted being considerably increased when the water 
intake was changed from 750 c.c. to 900 c.c. The subject had pre- 
viously excreted in the urine about 600 c.c. of water daily, but when 
the intake of water was increased, the amount given off in the urine was 
actually decreased for several days, so that an average of over 340 c.c. 
of water was retained per day for three days. This increase in the 
difference between the water taken and urine excreted would imply a 
distinct physiological need, since in the earlier experiments at Wesleyan 
University, in which the subjects fasted for a shorter period and the 
intake of water fluctuated widely, the variations in the amount of the 
urine followed very closely the variations in the amount of water 
ingested by the subject. 

Finally, it is significant that at no time was there any indication of a 
toxic effect in using distilled water, and we are able to sustain the con- 
tention of Winckler 1 that distilled water is without deleterious effect. 

^inckler, Zeitschr. f. diat. u. physikal. Therapie, 1905, 8, p. 567. 



BODY-TEMPERATURE. 

The profound alterations in metabolism in the body of a fasting man 
would lead one to expect some disturbance between thermogenesis and 
thermolysis. Body-temperature, which is the index of the resultant 
of these two factors, may obviously be affected by the disturbance of 
either. If there is a decrease in thermogenesis with no change in the 
thermolysis, there will be a fall in body-temperature. Conversely, if 
there is an increase in thermolysis with constancy of thermogenesis, 
there will again be a fall in temperature. 

In this laboratory body-temperature measurements have a dual 
significance : first, the value per se of the actual fluctuation, which indi- 
cates a disturbance in the relationship between thermolysis and thermo- 
genesis, and second, the importance of knowing body-temperature 
changes for the accurate computation of the heat production in the 
body. To determine the heat production it is not sufficient simply 
to measure the heat radiated from the body and to add to this value 
the heat of vaporization of water, for if in a given experimental period 
the body-temperature has decreased, there has been a loss of heat from 
the body unaccompanied by a production; hence the heat production 
is measured only after correcting for the body-temperature changes. 
In the series of body-temperature measurements in the short fasts at 
Wesleyan University, the average body-temperature did not alter 
noticeably, although there was distinct evidence of a flattening out of 
the curve showing the daily rhythm. In the prolonged fasting experi- 
ment with our subject L., we attempted to measure with the greatest 
degree of refinement all the factors. It seemed important, therefore, 
that frequent and careful records of the body-temperature should be 
made in connection with this fasting observation. 

The subject of body-temperature has been given special attention 
in this laboratory for a considerable period, and a year previous to this 
fasting experiment an extensive study of the fluctuations of the tem- 
perature in the various parts of the human body was reported. 1 As a 
result of this research, it became evident that the only suitable place 
for measuring body-temperature is deep in the body trunk, preferably 
in the rectum. The identical apparatus used in the study referred to 
was available for this fasting experiment and consequently body-tem- 
perature measurements were secured as frequently as possible. A 
detailed description of this apparatus and the tests made with it were 
published in the report cited. Briefly, the apparatus consists of a 
thermal element which is inserted about 7 cm. in the rectum, this 
thermal element being connected with another thermal junction in a 
constant-temperature bath. By means of this apparatus, it is possible 

Benedict and Slack, Carnegie Inst. Wash. Pub. No. 155, 1911. 
88 



BODY-TEMPERATURE. 89 

to determine the rectal temperature of a subject within 0.01°C, and 
records can be made as frequently as desired. 

Records of the rectal temperature were obtained nearly every night 
while the subject was in the calorimeter chamber, the junction being 
inserted in the rectum of the subject, connections made with the binding 
posts inside the chamber, and observations taken on an average of 
every 15 minutes throughout the night. On some nights records were 
taken every 5 or 6 minutes. Observations were also made at various 
times during the day and on at least two days continuous records were 
secured for nearly the whole day-period. The apparatus was frequently 
controlled by comparison with a standard thermometer, so we believe 
that these observations represent the absolute temperature changes 
of this individual. 

As in most of the measurements taken during the fast, the subject 
cooperated heartily in these body-temperature observations. After 
the first night, and, in fact, after the thermometer had been inserted 
a few moments, he experienced no particular difficulty and expressed 
himself as being very much pleased that this routine gave him no dis- 
comfort. It is clear that the use of the thermometer did not interfere 
in the slightest with his sleeping. The observations were therefore made 
under normal conditions, so far as it was possible to control them. 

The body-temperature measurements made in this fasting experi- 
ment may be considered in two ways: first, as to the alteration in the 
regular rhythm of the temperature as the fast progressed, and second, 
as to the effect of the fast upon the average of the temperature meas- 
urements. 

CHANGES IN TEMPERATURE RHYTHM. 

In order to study the first of these problems, namely, the changes in 
the temperature rhythm, curves have been plotted giving the tempera- 
ture values for the period beginning about 8 p.m. and ending about 10 
a. m. the following day. During this time the subject was in the 
calorimeter chamber from 8 p.m. until about 8 a.m., and then, without 
leaving the bed, he was withdrawn from the apparatus and was for the 
next two hours the subject of the morning respiration experiments. 
Accordingly, he was lying on the same bed in the same position for 
the entire time, the only change being that in the last two hours he 
was in the calorimeter room instead of inside the calorimeter chamber. 
Since the temperature of the calorimeter laboratory was essentially that 
of the respiration chamber, there was practically no alteration in the 
temperature environment throughout the whole period covered by the 
observations shown by the curve. 

It is, furthermore, of value to note that this period includes what is 
normally found to be the maximum diurnal change, for with normal 
individuals it has been shown that the lowest temperatures are found 
about 3 a. m. and the highest about 5 p. m. From 5 p. m. until 



90 



A STUDY OF PROLONGED FASTING. 



about 11 p. m., or until bedtime, the temperature usually remains 
approximately constant. The temperature, as a rule, falls rather 
rapidly after one goes to bed, reaching the minimum about 2 or 3 
a. m. With the fasting subject, the maximum temperature undoubt- 
edly was reached prior to his entering the chamber at 8 p. m., as he 
usually lay on the couch for an hour or more previous to that time. 
The body-temperature was unquestionably falling continuously during 
this preliminary period, so that the range for the night would be some- 
what less than the actual daily range. 



2:00 A.M. 4:00 



37.iT: 
369 

367 

363 

363 

361 

376 

37.4 

37.2 

37.0 

368 

366 

364 

362 

360 

37.1 

369 

367 

369 

363 

361 



v^ 


"©*' 


— •., 


■v 






















\ 






\ 


APR. 15- 
APR. 17- 
















18 




\© 


A 




\ 




























S ' 


"*\ 












/ 


















<N 








f t 


/ 




























.''A- 


S 






















*** 


M- 














\^ 






























































APR. 18 
APR. 19 
















ZO 




\® 


....... 
































.. ^ 




























N 


% * % 






























X ' 


"•v^ 




j/. 


























** — * 












\ 


<3L 


,^\ 






















































>® 




X 


APR. 2< 
APR.2 














22 










































•% 






.»*».. 


•«./' 






















2^zs> 















Fig. 4. — Body-temperature curves during the night and early morning 
for the second and fourth to eighth days of fast. 



BODY-TEMPERATURE . 



91 



The temperature curves for the period from 8 p. m. to about 10 a. m. 
for every day of the experiment, with but three exceptions, are given 
in figures 4 to 8. In order to save space, it has been necessary to plot 
the curves in pairs, but the observations are of such interest that it 
appears unwise to plot them in larger groups. The day of the fast is 
indicated by a number in a circle on each curve. It can be seen that the 
general trend of the curves remains essentially the same throughout the 
entire fast. There is a noticeable fall in the evening, the minimum 
being reached not far from 3 or 4 a. m. This is followed almost inva- 
riably by a distinct rise, which continues until the end of the record. 



e-.oo p.m. 



37.0°C 


*N, 


^ 






























A. 


r 


te« 


APR. 2; 
APR. 2: 


r 












j~ 


36.8 




36.6 












X 












,.-'' 






36.4 














\ 






..-"•-. 


/ 








36.2 






























37.3 




(0 






























5jr 


Z\, 




APR. Zt 
APR. 2! 


1-25 










37.1 




36.9 








■^x 


^S. 




















36.7 


















A 


./\ 


'{"" 


.-— '' 






36.5 
























rW 






37.0 




Lp> 


/ 


v 


























\® 






y\ APR. 2( 
\ APR. 2" 














! 


36.8 




36.6 




• 


V 
























36.4 








*•* 


^ 














/ 






36.2 












\ 


— -. * 






j — X 


r*-*». 


--'' 






36.0 
















\ 


-..--' 


/ 










37.0 


\ 


.,© 


























36.8 




# 


5. 
































^v 


APR. 2 
APR. 2 
















36.6 




36.4 










nN 


. 


















36.2 












s^ 








-rS^ 


/ s 


^.♦s^ 






36.0 































Fig. 5. — Body-temperature curves during the night and early morning 
for the ninth to the sixteenth days of fast. 



92 



A STUDY OF PROLONGED FASTING. 



OBSERVATIONS OF THE BODY-TEMPERATURE IN THE NIGHT PERIOD. 
AVERAGE BODY-TEMPERATURE. 

The greatest interest, at least to the clinician, lies not in the course 
of the temperature curve throughout the night, but in the average tem- 
perature values as the fast progresses. These are recorded in table 
5 (page 95) for nearly every night of the fasting experiment, only the 
time that the subject was inside the bed calorimeter being included in the 
values. These averages were taken directly from plotted curves and, 
except in the values for the tenth to the fourteenth nights of the fast. 



37.0C 




v£2> 


























368 




s - 
































\ 












, 




MAY 1-2 








366 




364 








\ 


















1 '• 
/ 












\ 












^/" 






t 




If.? 






360 














V" 




****** 












.WO 


>j, 


































M 




MAY 2 
MAY 3 


_ 








; 


' 






368 




Ififi 






\© 


\.. 


^ 
















J 


fS 


VI 4 












\ 


.. 






, 


...-•' 


1 


\f 




36? 














\ 




*•%.--"* 


»"* 










360 






























Mil 






























369 




N^ 
























• 


.1*7 




@ X - 


. \ 
























361 






\ 


















. 


1 


\/" 








\ 




MAY 4 
MAY 5 


_ 






fj 










363 




361 








V « 














•' 








319 








V 


A / 






— > 














357 










\ 


\, 


















35.5 














V" 

















Fig. 6. — Body-temperature curves during the night and early morning 
for the seventeenth to twenty-second days of fast. 



BODY-TEMPERATURE . 



93 



show a general tendency for the temperature to remain reasonably- 
constant up to the seventeenth night of fasting. The temperature then 
fluctuated, falling as low as 35.88° C. on the twenty-fourth night of 
fasting and rising as high as 36.37° C. on the twenty-eighth night of the 
fast. The maximum average value for the body-temperature observed 
in any night during the fast was 36.85° C. on the twelfth night and the 
minimum value was 35.88° C. on the twenty-fourth night. On the 
last night of the fast, the average body-temperature was 36.14° C. 



8:00RM. 


10:00 


12:00 


2:00AM. 


4:00 


6:00 


8:00 


10:00 


36.5°C 


































**„ 


**-.-'' 


^ 




MAY 6 
MAY 7 


-7 














36.3 






36.1 






^ 


\ 




















\/, 




35.9 








V 


-■ 














'"/ 








35.7 










•• \f 




V w 




— "s/" 


*w 




<s 








37.2 
































37 


/* 




\© 


























36.8 










































MAY 8 
MAY 10 


n 










/ 










.16 6 






36 4 




-"\.'\ 






W 


\ 




















36.2 






©N 






Vn 




















36.0 












"•—. 


-Vf 


\ 










t 


' 




35.8 
















»-^*- 




V. 


,-*"' 


V 








368 






















\J 












""""' 


\ f 


\p> 






MAY II 
MAY 12 








/ 


y f 










36.6 






36.4 






@^ 


/ V 


. ( 


















y4 

t 

1 




36.2 










^ 


■ 




















36.0 












\.., 


../ 


"V/ 


"**\ 


" V 




r 








35.8 






















w v 


t 









Fio. 7. — Body-temperature curves during the night and early morning for twenty-third to 

twenty-ninth days of fast. 



94 



A STUDY OF PROLONGED FASTING. 



With the resumption of food the average temperature increased on 
May 16-17 to 36.79° C. and on May 17-18 to 37.53° C. 

RANGE IN BODY-TEMPERATURE. 

The maximum and minimum temperatures and the range in the 
temperature for each night are likewise recorded in table 5. The maxi- 
mum temperature observed on any fasting night was 37.45° C. on the 
fifth night; the minimum temperature was 35.61° C. on the twenty- 
second and twenty-third nights. The difference between the minimum 
and maximum values, or the range in temperature, is also recorded in 
table 5. The average range was not far from 0.90° C. The maximum 
range, 1.27° C, was observed on the fifth night of fasting; the minimum 



36.7 C 

36.5 

36.3 

36.1 

35.9 

35.7 



37.1 
36.9 
36.7 
96.5 

37.9 
377 
37.5 
37.3 



V 


— *"""*> 




























N@, 






MAY 13 
MAY 14 






















\ 


*^^w. *' 


*\ 


-\ 














/' 


•• 














— \ 










i 




















Li -V 
































^jS' 


• ** 

V 
































































v 


^o . 






































MAY 16-17 


































































































V 






























% 


*\ 








MAY 17-18 


f 


>"*N 
















A 


fS 


/- 


V*-»i 


/ 


/ 


















V 


V 


^ 


"^\ 


U^ 















Fig. 8. — Body-temperature curves during the night and early morning for thirtieth and thirty-first 
days of fast and second and third days with food. 



BODY-TEMPERATURE . 



95 



range was 0.50° C. on the twelfth night of the fast. It is important to 
note, however, that the variations in the range have not even the 
semblance of regularity. 

OBSERVATIONS OF THE BODY-TEMPERATURE IN THE DAY PERIOD. 

On two days (May 7-8 and May 8-9) the body-temperature was 
measured almost continuously throughout the entire 24 hours. Curves 
showing the fluctuations in body-temperature on these days are given 

Table 5. — Body-temperature (rectal) of subject L. during experiments in the bed 

calorimeter at night. 



Date. 


Day of 
fast. 1 


Max. 


Min. 


Range. 


Average. 


Observation 
at 7 a.m. 


1912. 




°C. 


°C. 


°C. 


°C. 


°C. 


Apr. 15-16 


2d 


36.92 


35.97 


0.95 


36.41 


36.44 


17-18 


4th 


37.23 


36.01 


1.22 


36.55 


36.36 


18-19 


5th 


37.45 


36.18 


1.27 


36.58 


36.50 


19-20 


6th 


36.85 


36.14 


.71 


36.44 


36.42 


20-21 


7th. . . . 


36.74 


36.18 


.56 


36.42 


36.28 


21-22 


8th 


37.17 


36.25 


.92 


36.55 


36.53 


22-23 


9th 


37.12 


36.11 


1.01 


36.50 


36.41 


23-24 


10th 


36.96 


36.39 


.57 


36.64 


36.66 


24-25 


11th.... 


37.28 


36.52 


.76 


36.80 


36.54 


25-26 


12th 


37.19 


36.69 


.50 


36.85 


36.81 


26-27 


13th 


37.04 


36.13 


.91 


36.62 


36.63 


27-28 


14th 


36.60 


35.99 


.61 


36.30 


36.28 


28-29 


15th 


36.95 


36.11 


.84 


36.43 


36.54 


29-30 


16th 


36.92 


36.16 


.76 


36.40 


36.31 


Apr. 30-May 1... 


17th.... 


37.05 


36.04 


1.01 


36.42 


36.45 


May 1-2 


18th.... 


36.95 


35.97 


.98 


36.30 


36.31 


2-3 


19th 


36.86 


35.82 


1.04 


36.21 


36.40 


3-4 


20th 


36.88 


36.19 


.69 


36.51 


36.61 


4-5 


21st 


36.96 


35.70 


1.26 


36.12 


36.04 


5-6 


22d 


36.67 


35.61 


1.06 


36.10 


36.31 


6-7 


23d 


36.48 


35.61 


.87 


35.98 


35.84 


7-8 


24th 


36.38 


35.68 


.70 


35.88 


35.78 


8-9 


25th 


37.00 


35.99 


1.01 


36.31 


36.36 


10-11 


27th 


36.27 


35.76 


.51 


36.03 


35.98 


11-12 


28th 


36.69 


36.04 


.65 


36.37 


36.10 


12-13 


29th 


36.75 


35.92 


.83 


36.23 


35.92 


13-14 


30th 


36.57 


35.79 


.78 


36.06 


35.94 


14-15 


31st 


36.74 


35.74 


1.00 


36.14 


35.96 


16-17 




37.15 
37.78 


36.50 
37.40 


.65 
.38 


36.79 
37.53 


36.58 
•37.68 


17-18* 









'For the duration of the period during which these observations were made, see table 44. 
2 The maximum temperature on this night was observed near the end of the calorimeter period. 
•This observation was obtained during the morning respiration experiment, the subject lying 
on the couch after leaving the calorimeter. 

in figure 9. These curves, which were obtained on the twenty-fourth 
and twenty-fifth days of the fasting experiment, show that even late 
in the fast there was a very large diurnal variation. 

It has been stated that the range in temperature during the night 
was not a criterion of the probable total range throughout the 24-hour 



96 



A STUDY OF PROLONGED FASTING. 



day, for before the night observation began the subject had been in a 
condition of rest for one or more hours. On these two days the maxi- 
mum temperature observations occurred during the daytime, as, for 
instance, on May 7-8, when the maximum temperature observed was 
37.10° C. at about 5 p. m., the normal hour of the day. The minimum 
record was 35.68° C. at 4 a. m., the entire range being 1.42° C, a value 
exceeding any range given in table 5. Similarly on May 8-9, the high- 
est temperature recorded was in the daytime at about 12 h 15 m p. m., 
when a temperature of 37.51° C. was recorded. The minimum value 
was 35.99° C, at 2 h 40 m a. m. Thus the range was 1.52° C, exceeding 
even that of the preceding day. These curves give a general picture 
of what was probably the average daily course of the body-temperature 



BOOPM 


10 00 


K 


00 


2:00 AM. 


4:00 


6.00 


8:00 


I0O0 


12:00 


2:00 P.M. 


4:00 


e.< 


T7-4.V. 
































, / 


/•V-^ 












37? 






























f 


y 

r 






V^\/' 


/ 


'X 




370 


** 




\© 
























I 






















\ 






MAV 7 

MAV 8 


_ 










J 


? 






*\ 
















36.8 






36.6 








^ 




















/ 








\ 










36.4 










\ ^-. 














S 






















36.2 






@\ 






\r-. 










/ 
























360 














• l 


sy - ' 




1 


~/ 




r 




















35.8 
























/ 






















356 
























J 























Flo. 9. — Body-temperature curves for approximately 24 hours on twenty-fourth and twenty-fifth days of fast. 



of this subject throughout the 31 days of fasting. Here, again, there is a 
distinct tendency for the maximum temperature to appear in the late 
afternoon and the minimum temperature in the early morning, this not 
being affected by many days of fasting. 

On three other days, body-temperature records were obtained for a 
part of the morning. On May 1 and 4, observations were made during 
respiration experiments in which L. was sitting up and writing. (See 
plate 1, fig. B.) These values are given in figure 1 0. This experiment 
followed the regular morning respiration experiment, in which the 
subject lay upon a couch; the observations recorded for the lying 
position are also given in figure 10 for comparison. Of particular 
interest is the fact that the change in position from lying to sitting 
did not greatly alter the rate of the morning rise in temperature. 

On May 15, when the subject first began to eat, the observations 
commenced shortly after the end of the regular respiration experiment 



BODY-TEMPERATURE. 



97 



and continued until noon. During this time the subject was sitting up 
and eating. The curve given in figure 11 for both the respiration 
experiment and the eating period shows that when the subject was 
sitting and eating the ascent is somewhat more noticeable than in the 
lying period, but it is evident that even eating after a 31-day fast did 
not materially disturb the course of the rectal temperature curve. 

CONSTANCY IN BODY-TEMPERATURE AT A GIVEN HOUR. 

Since at 7 a. m. the subject had been living under constant conditions 
of quiet and rest inside the chamber for 8 or 10 hours, a comparison 
may be made of the values for the body-temperature obtained at this 
time from day to day. This comparison is the more important since 



373 o c 8:ooa.m 



37.1 s 



9:00 



10:00 



36.9" 



3G.7 



36.5 



MAY 1 
MAY 4 














J>, 


• 

1 






/ 


• 








/ / 
/ / 
/ / 

f 











Fig. 10. — Body-temperature curves showing change from lying to 
sitting position. 



37.(f c 8:00 A.M. 



36.8* 



36.6° 



36.2° 




1200 



Fig. 11. — Body-temperature curve showing change from lying to 
sitting position. 



98 A STUDY OF PROLONGED FASTING. 

in many fasts the body-temperature measurements are made but once 
each day and usually at a given hour. Accordingly, in table 5 the 
temperature records obtained at or near 7 a. m. have been given for 
each day. In general, the variations in the temperature at 7 a. m. are 
not markedly different from the variations in the average temperature 
throughout the night, since the maximum and minimum records for 
this time were found on the same days as the maximum and minimum 
average temperatures. The maximum value found at 7 a. m. was 
36.81° C, at the end of the twelfth night of fasting, and the minimum 
value 35.78° C, at the end of the twenty-fourth night of fasting. No 
uniformity in the values is apparent. 

With the fluctuations in the body-temperature varying as they do 
it will be seen that the difficulties in securing an average temperature 
throughout the fasting period by means of one or two observations 
during the day have been overlooked. Only by securing average values 
throughout the entire night or throughout several hours at approxi- 
mately the same time each day can a true picture of the average temper- 
ature change of the body as affected by inanition be secured. 

The well-known influence of muscular activity on body-temperature 
makes it the more regrettable that certain experiments with muscular 
activity could not have been carried out with this subject, as the effect 
of a moderate amount of muscular exercise upon the temperature regu- 
lation as the fast progressed would have great theoretical interest. 
This is one of the problems that should certainly be studied in any 
subsequent fasting experiment. 

The observations of body-temperature on other fasting individuals 
have frequently been made without reference to the preceding muscular 
activity or the general condition of the sub j ect . Obviously those made in 
the morning, just before the subject rises, have by far the greatest value. 
It is a characteristic of practically all the fasts heretofore reported — in 
which the temperature observations have been made for the most part 
in the axilla or in the mouth (both localities unsuited for physiological 
experiments) — that there has not been sufficient disturbance in the 
temperature regulation to be recorded by this method of thermometry. 



PULSE-RATE. 

In practically all of the fasting experiments with which we are 
familiar, the method of taking the pulse-rate from the radial artery has 
been used. In the fasting experiments made at Wesleyan University, 
in which the subject remained in the calorimeter during the whole 
period, it was at first necessary to rely upon the subject's own obser- 
vations of the radial pulse. This method was by no means ideal and, 
in a later series of 2-day fasting experiments with seven individuals, 
the method was improved upon, in that a small tube-pneumograph was 
placed about the chest. The pulse-beats were thus superimposed upon 
the respiration movements of the tambour and could be counted by an 
observer outside of the chamber. 

The striking relationship between pulse-rate and metabolism, which 
has been regularly noted in this laboratory for many years, not only 
with men but with animals and more recently with infants, led us to be 
especially interested in the pulse-rate of our fasting subject. For a 
study of the pulse-rate during the fasting experiments, it was necessary 
to select a method by which continuous records could be made, as the 
pulse-rate gives a reasonably accurate index of the metabolism at the 
time the pulse record is made. The method of recording the pulse-rate 
from the radial artery, either by an observer or by the subject himself, 
has distinct disadvantages in that the knowledge that the observation 
is being made has a psychical influence which is undesirable. Con- 
tinuous records, therefore, could not be obtained by this method. 
Furthermore, while the pneumograph method may properly be used 
in a short experiment, its use in a long-continued experiment is objec- 
tionable. The wearing of the pneumograph for a considerable period 
of time may cause the subject much discomfort, as the traction becomes 
wearisome, and if he changes his position during the experiment inside 
the calorimeter the pneumograph may possibly press into the flesh and 
be somewhat painful. The transmission tube may also become twisted 
and thus interrupt the record. 

It was hoped that this continuous record of the pulse-rate could be 
obtained by photographic registration with the string galvanometer, 
but although Professor W. B. Cannon kindly loaned us the string gal- 
vanometer belonging to the Harvard Medical School, it was impossible 
to install and test it suitably in time for its use in this experiment. 
In our previous experimenting we had found it advantageous to fasten 
the bell of a Bowles stethoscope over the apex beat of the heart and by 
using long transmission tubes very satisfactory counts of the pulse-rate 
were obtained. Accordingly, since records could not be made by 
photographic registration, the stethoscope was used for nearly all of the 
observations in this fasting experiment. The stethoscope is much less 

99 



100 A STUDY OF PROLONGED FASTING. 

disturbing to the subject than feeling of the radial pulse, but a few 
additional records were obtained by the latter method. In the later 
days of the fast, when the apex beat of the heart became fainter, it was 
occasionally necessary for the observer in the respiration experiments 
to note the pulsations of the carotid artery. 

The pulse-rate records may be classed in two groups. The first in- 
cludes a large number of perfectly comparable observations : those made 
throughout the night, while the subject was inside the bed calorimeter, 
and those during the 1§ or 2 hours of the morning respiration experiment. 
Usually the period of continuous observation extended from 8 p. m. to 
9 h 30 m or 10 a. m. ; during this time the subject was lying quietly upon 
a couch. These records were made regularly every day of the fast. 

While the subject was in the bed calorimeter, the records were made 
by the regular chemical assistant as often as possible, the frequency of 
the observations obviously depending somewhat upon his other duties. 
Occasionally when the subject moved inside the calorimeter, so as to 
slightly displace the bell of the stethoscope, the pulse beats could not 
be heard and there would consequently be a break in the records until 
the subject again changed his position so as to bring the bell to its 
former location. During the morning respiration experiment a special 
observer was detailed to count the pulse-rate continuously throughout 
the whole period. (See plate 2, figure C, page 19). 

The second group of observations consists of those taken at various 
times throughout the day, a part of which were continuous, while others 
were individual records. This group includes the observations in the 
miscellaneous respiration experiments, such as those made in the even- 
ing before the subject entered the bed calorimeter, while the subject was 
writing, or when he was breathing an oxygen-rich atmosphere. During 
the latter part of the fast, the pulse-rate was also recorded twice when 
the daily record of the blood pressure was taken, and occasionally when 
other special tests were made. At times the subject wore the stetho- 
scope throughout the whole day, so that the observations were more or 
less continuous for the 24 hours. On the days when the continuous 
observations were made, the subject was followed by an assistant who 
kept out of sight but made the records regularly and also noted the 
changes in body-position. These records were frequently verified by 
a second observer. 

RECORDS OF PULSE-RATE OBTAINED IN EARLIER FASTING EXPERIMENTS. 

Before giving the records of the pulse-rate obtained in the fasting 
experiment with L., it will be of interest to cite those secured in fasting 
experiments made by other investigators. In discussing such obser- 
vations, two essentially different comparisons can be made, first, the 
influence of a prolonged fast upon the pulse-rate determined under any 
given conditions, and second, the variations in pulse-rate incidental to 
the changes in position or mental activity. Usually in fasting experi- 



PULSE-RATE. 101 

ments observers have contented themselves with taking the morning 
pulse-rate and occasionally the evening pulse-rate. No particular em- 
phasis has been placed upon these individual observations, aside from 
the general fact that the pulse may have altered as the fast progressed. 
Not recognizing the great significance of the pulse-rate in relation to the 
metabolism, experimenters have not ordinarily taken especial precau- 
tions (as did Luciani) to keep the suoject lying quietly while the pulse- 
rate was being observed and, indeed, for some time previous to the 
observation. This probably explains difficulties found in comparing 
the records, in that some observers note a continually decreasing pulse- 
rate during the fast, while others find marked irregularities. As would 
be expected, the more recent observations take into account the factors 
influencing the pulse-rate and the records are thus more trustworthy. 

Of the pulse records obtained in Tanner's fast, we have been able to 
find only those given in the British Medical Journal. 1 On the thirty- 
seventh day of this fast, the pulse, respiration, and temperature are 
reported as having been "normal." On the twenty-fifth day the pulse- 
rate is given as 75, the respiration as 15, and the temperature of the 
mouth as 98.4° F. (36.89° C). On the thirtieth day the pulse-rate was 
reported as 84 and slightly more regular, the temperature as 98.8° F. 
(37.11° C), and the respiration as 14, with the general statement that 
he was weaker than on any previous day. "On the twenty-ninth day, 
two of the experts attending him reported that there was no material 
alteration in the vascular pressure indicated by the heart's impulse, 
while its volume was scarcely less than in health." 

Paton and Stockman 2 report that the pulse-rate of their subject 
averaged between 50 and 60 and the respiration usually between 23 and 
30, but no continuous records of the pulse-rate are given. 

The most extensive series of continuous observations of the pulse- 
rate of a fasting subject is that reported by Hoover and Sollmann. 3 
In this 5-day fast, the pulse was counted and recorded once every hour 
by relays of watchers. The initial record of the pulse-rate was 75, the 
lowest value of 37 being recorded on the last day, thus showing a dis- 
tinct tendency for the pulse-rate to decrease as the fast progressed. 
Unfortunately the fast continued for only 5 days and, in the opinion 
of the authors, the pulse records are vitiated by the fact that they were 
obtained with a hypnotic subject and that the pulse-rate was purposely 
lowered by suggestion. 

In reporting a fast carried out by Succi in New York in December 
1890, and said to have continued for 45 days, a newspaper states 4 that 
on the last day of the fast Succi's pulse-rate was 62. Unfortunately no 
scientific record of this fast was ever published. 

British Med. Journ., 1880, 2, p. 171. 

2 Paton and Stockman, Proo. Roy. Soc, Edinburgh, 1888-1889, 16, p. 121. 

'Hoover and Sollmann, Journ. Exp. Med., 1897, 2, p. 403. 

«N. Y. Daily Tribune, December 21, 1890. 



102 A STUDY OF PROLONGED FASTING. 

In a fast carried out by Succi in London in 1890, which continued 
40 days, the pulse-rates, taken every day at noon, 1 varied from 82 on 
the second day of the fast to 52 on the thirty-fifth day. The degree of 
irregularity noted in all conditions of the fast, however, shows that 
proper attention had not been paid to secure uniform quiet before the 
observations were made. The respirations varied from 16 on the 
thirty-first day of the fast to 28 on the sixth day of the fast. Here 
again the irregularity noted on all days implies variations in muscular 
activity prior to the observations, no general trend of the respiration 
rate being apparent. 

In the 10-day fasting experiment with Cetti, 2 the pulse-rate ranged 
from 68 on the morning of the fourth day to 92 in the afternoon of the 
seventh day. The high pulse-rate was accompanied by abdominal 
pains. In certain of the respiration experiments carried out with Cetti, 
the pu]se-rate was likewise recorded. In one instance it was noted that 
the pulse-rate changed from 88 while the subject was lying down to 120 
while he was walking about the room. In another experiment the 
pulse-rate changed from 86 while he was lying down to 98 when he was 
sitting, smoking, and talking. The great increase in the heart action 
of this subject was commented on at some length by these authors. 

In an experiment with Breithaupt, continuing for 6 days, the same 
authors record a minimum pulse-rate of 47 on the last day of the fast 
and a maximum rate of 66 on the morning of the second day. Taking 
advantage of the fact that their subject performed muscular work on 
the ergostat, the authors made some interesting notes upon the increase 
in the heart-beat with a definite amount of work and the return of the 
pulse-rate to normal after the work ceased. In their general conclusions 
they maintained that with Cetti, who was of an excitable temperament, 
the pulse-rate in the resting condition was not noticeably changed by 
fasting, but that it slowly decreased with Breithaupt, who was quiet 
and phlegmatic. They also emphasize the fact that during the fast 
there was a distinct tendency to a considerable increase in the irri- 
tability of the heart, slight muscular activity producing a great increase 
in the pulse-rate. 

Luciani contends that, during his experiment with Succi, the pulse- 
rate remained strictly inside the physiological limits, rising to 70 but 
twice and only occasionally falling below 50. He also points out that 
the pulse-rate, as well as the temperature and the respiration, were 
always measured during complete muscular rest, as the subject was 
lying in bed. An interesting observation on the irritability of the heart, 
as indicated by the rise in the pulse-rate after exercise, was likewise 
made by Luciani, who was fortunate in having a fasting subject who 
freely indulged in muscular activity. 

British Medical Journal, 1890, pp. 764, 819, 876, 935, 996, 1056, and 1444. 
*Lehmann, Mueller, Munk, Senator, and Zuntz, Archiv f. path. Anat. u. Physiol, u. f. klin. Med., 
1893, 131, Supp., p. 1. 



PULSE-EATE. 103 

The pulse-rates were also recorded in a long fast made by Penny. 1 
This fast was less strictly controlled than the previous fasts cited, 
but Penny states that his observations of the pulse-rate were verified by 
another doctor. The morning observations were made about 9 o'clock 
before he rose from his bed; the evening pulse-rate was taken about the 
time of retiring, i. e., 10 or 11 o'clock. The records for the morning 
ranged from 59 on the second day of the fast to 39 on the thirteenth, 
fourteenth, fifteenth, and eighteenth days of the fast. The evening 
records ranged from 80 on the last day to 44 on the eleventh, fifteenth, 
and sixteenth days of the fast. 

Far less confidence can be placed in the observations reported for 
Gayer, who was said to have carried out a 30-day fast in New York in 
1910. My only justification for calling attention to these observations 
in the report of this fast is the personal assurance of Dr. Ira S. Wile, 
of New York, who, while not vouching for the authenticity of the fast, 
is inclined to believe that the records are for the most part trustworthy. 
These show a pulse-rate ranging from 54 to 80, but, as the writer points 
out, the maximum observation was taken after the subject had come in 
from a 2-mile walk and on the very next day a pulse-rate of 61 was 
noted when the subject spent the morning lying down. 

Cathcart 2 recorded both the morning and evening pulse-rates of his 
subject, Beaute. Charteris 3 also recorded the pulse-rates on this indi- 
vidual, but obviously at a slightly different time of day, as his records 
do not agree with those of Cathcart. Nevertheless both authors draw 
the conclusion that there was a general tendency for the pulse-rate to 
fall as the fast progressed. Charteris furthermore points out that the 
subject was well aware of this fact from his previous experience, as he 
was a professional faster. Cathcart's morning observations ranged from 

70 on the seventh day of the fast to 58 on the twelfth and fourteenth 
days of the fast. The highest observation secured in the evening was 

71 on the second day, and the lowest was 57 on the tenth day of fast- 
ing. The records obtained by Charteris show a range from 68 on the 
second day of fasting to 58 on the twelfth day of the fast. 

RECORDS OF PULSE-RATE OBTAINED IN THE EXPERIMENT WITH SUBJECT L. 

The number of observations obtained with L. was sufficient to justify 
their presentation in the form of 24-hour curves, as shown in figures 12 
to 18. In these curves the day begins with 8 p. m., when the subject 
entered the respiration calorimeter, ending 24 hours later. Continuous 
records were secured for every night experiment and frequent records 
were made during the day. The values are perfectly comparable for 
each day between 8 p. m. and 10 a. m. and also for the most part 
throughout the rest of the day, as the daily routine of the subject was 

^enny, British Med. Journ., 1909, p. 1414. 
"Cathcart, Biochem., Zeitschr., 1907, 6, p. 109. 
J Charteris, Lancet, 1907, 173, p. 685. 



104 



A STUDY OF PROLONGED FASTING. 



more or less regular. The pulse records which were obtained in the 
evening respiration experiments may logically be attached to the 
records for the bed-calorimeter experiments as preliminary periods, but 
the fact that the evening experiments were made only in the latter 
part of the fast complicated their presentation in this manner. As a 
matter of fact, the record of the pulse-rate from the beginning of the 
evening respiration experiment, i. e., about 7 p. m., until the close of 



100 



8:00 P.M. 10:00 



12:00 



2:00A.M. 4:00 



6:00 



8:00 



10:00 




Fig. 12. — Pulse-rate chart of subject L. for days preceding fast. 



PULSE-RATE. 



105 



the respiration experiment the next morning at 9 h 30 m or 10 a. m., was 
continuous, as the subject did not rise from his couch during that period. 
He urinated lying on the side. Since the conditions of activity were 
essentially uniform from the time the subject entered the bed calorimeter 
at about 8 p. m. until the end of the morning respiration experiment 
at 9 h 30 m or 10 a. m., the records of the pulse-rate taken during this 
period on every day of the fast are more comparable. It is therefore 



) P.M. 10:00 IfcQO 2:00A.M. 4:00 8:00 8:00 10:00 12:00 2:00 P.M. 4:00 6:00 8:00 



H£- 



A.x,\www 



o: 



^ 



k 







4^\ 



fiL 



V 



\ 



1 





V 



® 



Fio. 13. — Pulse-rate chart of subject L. for first to fifth days of fast. 



106 



A STUDY OF PROLONGED FASTING. 



permissible to discuss these observations first, and later consider the 
more or less heterogeneous observations taken during the day when the 
activity might vary. 

PULSE-RATE IN THE NIGHT PERIODS. 

The relatively large fluctuations in the pulse-rate that are apparent 
in the first two or three nights are naturally to be explained by the fact 
that the subject was a stranger in America, and was experiencing for 



if 


JOPU. w 


00 12: 


X) M 


0AM «■ 


» «( 


10 S: 


30 W: 


>0 tt: 


90 1:C 


K)».M. 4. 


30 


6. 


X) 8.001 


T«i 




















J 


\h 






M 




\®. 


1 


pa. i»- 


M 




flju 




■ 






J* 


H 














V* 










\J 


to 




























PO 




























m 
















A 


Va 


. 


j 








y^~ 


^, 


A 


m jo-j 


1 


,l 1 


V^ 


' V 


f 












y 


-^ n 




^X 


1 














ri 






M 








APR. 21 


■22 




L /^ 












ss 




<sj^ 








\rS-J 


v 












7(1 


h 
























fiO 


*\ 


*v 


A 


PR. 22- 

A 7\ 


23 


N 


»y 




- 




■ 


^*- 


•XI 




vy 


V 


AJW| 


W) 


nji 














tl) 




















l 


A 




«0 


/> 


© 


APR. 


23-24 


» 


\k 


i/^ 




. 


1 


1^ 




so 




V 


*"N 


vAV 


y^J 


VVy 














N 












f\ 














H 




(£V 




APR. 2 


♦ -25 




^ 












4S 



























Fig. 14. — Pulse-rate chart of subject L. for sixth to 
eleventh days of fast. 



PULSE-RATE. 



107 



the first time the novel sensation of being inclosed in the respiration 
chamber. On the night of April 14-15 (figure 13), we find reasonably- 
constant pulse-rates until 4 a. m., when the observer's records show 
that he woke up, then dozed for the rest of the night. Usually the 
pulse-rate showed a tendency to fall prior to midnight, thereafter to 
continue fairly low until it rose again in the morning, although the 
period of minimum pulse-rate might continue for several hours. As 
the fast progressed, there was a marked tendency for the amplitude of 



^8 <X)P.M.IO:00 12:00 2:00A.M. 4:00 6:00 8:00 10:00 12.00 2:00P.M. AM 6:00 Ofl O 



60 


\ 


"\ 


1 

APR 2 


5-26 


/\l\ 


. l 


ij/t 






Ijl 


A 




SO 




dr 


K 


Ja 


fi\ 


jj 


7 






Ir- 




\ 


u 


















^ 








H 


^V 


^v 


APR. 


26 -Zf 


-1* 


\ K 


*/ 




r 


#* 






7,1 








\_j-r°r 


SJ 


W* 














65 
















/v 










is 


rv 


^© 


APR. 


27-28 




tA 


/W/rV 




- 






r 


45 




1 


"V 


t* 


y 


4\s* 














70 






















1 




*0 


















• 


/i 




p 


SO 




d\ 




APR. 21 


1-29 


jJ 


V 










I 


60 
























/v 


SO 




©\ 


1 


kPR. 29- 


30 

_A-- 


K-J 


W* 










r 


70 
















\ 


\ 








60 














J 


V 


\ 






\ ,/• 


10 






PR. 30 

l L 


-MAV 1 

A_ 




^J 


V 










/ 


fil 










—rb 
















IS 


S 


-y$2> 




«av i- 


2 




Ar 




„ 






/ 




























45 







Fig. 15. — Pulse-rate chart of subject L. for twelfth to 
eighteenth days of fast. 



108 



A STUDY OF PROLONGED FASTING. 



the curve to become less and less — that is, the fluctuations from maxi- 
mum to minimum throughout the night were less and the periods of 
reasonably constant pulse values grew longer and longer. 

On the night of May 14-15, the last night of the fast (figure 18), 
special attention was given to the pulse-rate, the records being made 
frequently throughout the whole night. Although the curve is in con- 

Jjgjl 10:00 tt-OO MO*-* +OQ fcOO fcOO 10-00 IfcOO jjOMj, 4c00 6:00 8:00 



— "I 












1 




\ 








■a 


©V^ 




HAY 2- 


3 


N 


v* 










'1 


» 
























«nL 


1 


IAY i-4 








,v 


i 


\ 








■a 


A@L 


JU 


-A. 


/ 


^i 


w* 












«s 
























w\ 


*t 


MAY 


*-s 




Hi 


iH 




X 






J 1 


ta 


V 


MJ 


u 


A^ 


IH, 


/ 










'V 


















I 








« 


Ji© 




HAY 5- 


« 


./ 


/ 




1 






/* 


4S 








r* 


.-vA-* 














TD 


















1 


1 




-A 

Ml 


& 


Mj 


lY 6-7 




A 1 


i 




I 


1 


1 


/ 


SO 


\ 


W. 


a 


w^> 


V 


uw 










l> 


■K 
























M 
























-v\ 


n® 


MA 


r- 7-8 


A /-l 


}j 


\/ 










** 


«n 


V-\ 


V^-v 


^V^\ 


y^ 1 


~J U 














«n ~ 


^ 


1 


HAY 8- 


i 








1 






/ 


■I 


\A. 


\J 


^ 


v\ 


JW 


V* 1 










1 



Fia. 16. — Pulse-rate chart of subject L. for nineteenth to twenty-fifth days of fast. 



PULSE-KATE. 



109 



sequence irregular in shape, the general trend is not markedly different 
from those for the preceding and following nights. Even on May 15-16, 
the first night following the ingestion of food, although the subject was 
in such distress that he did not go inside the chamber, but lay on a couch 
outside, the frequent records of the pulse-rate did not show extraordi- 
narily large fluctuations. On the night of May 16-17 relatively few 
records of the pulse-rate were taken; and also on May 17-18, but on 
this night we find a greatly increased amplitude. The general deduc- 
tion is, therefore, that the amplitude of the fluctuations of the pulse- 
rate during the night decreased regularly as the fast progressed, showing 
a tendency upon the resumption of feeding to return to the variations 
commonly experienced. 



00P.M. 10:00 12:00 2:00A.M. 4:00 6:00 8:00 10:00 12:00 2:00P,M. 4:00 6:00 8:00 



60 



50 




^N^AjA/^' 




Fig. 17. — Pulse-rate chart of subject L. for twenty-sixth to thirtieth days of fast. 



110 



A STUDY OF PROLONGED FASTING. 



PULSE-RATE IN THE DAY PERIODS. 



From 10 a. m. until 7 p. m., the records are naturally much less com- 
plete than the series obtained during the night; nevertheless on certain 
days reasonably complete records of the pulse-rate were obtained 
throughout the day. 



8:00 P.M KMX) 12.00 200 AM 4:00 6:00 8:00 KMW IZ.00 2:00A.M. 4:00 6:00 8:00 




Fiq. 18. — Pulse-rate chart of subject L. for thirty-first day of fast and three subsequent days 
with food. The point when the subject took food on May 14-15 is indicated by a heavy 
vertical line. 



PULSE-RATE. Ill 

On April 16-17 (figure 13), the records were made for nearly the whole 
day, these values probably being fairly typical of records which would 
have been obtained if the observations had been more complete on other 
days. The minor fluctuations shown on April 16-17 are obviously due 
to changes in the activity of the subject when talking or moving about. 
The high values obtained about 5 p. m. are coincidental with the hand 
dynamometer test, in which there was some muscular exertion by the 
subject, the highest record at this time being 102. After the dyna- 
mometer test was over, the pulse-rate immediately fell again to an 
approximately normal level. Until 5 h 30 m p. m., the subject was lying 
on a couch and from 6 h 05 m p. m. was asleep in his chair for half an 
hour, the low level of the pulse-rate being apparent at this time. The 
records shown by the last portion of the curve were obtained during 
the evening respiration experiment. For the greater part of this day, 
the pulse-rate was on the average not far from 10 beats per minute 
above that which would ordinarily be found when the subject was lying 
upon a couch, although during muscular exertion, and especially after 
the dynamometer test, the pulse-rate at times tended to rise consider- 
ably above this value. The increase as a result of the dynamometer 
test may also be noted on April 18-19, April 19-20, and April 20-21. 

Beginning with April 30-May 1, records were made each day not 
far from 1 p. m., at the time of the blood-pressure test. These records 
are of unusual interest, inasmuch as they indicate the values while 
the subject was sitting and again immediately afterwards when he lay 
down upon the couch. Thus, on April 30-May 1 (figure 15) the record 
for the sitting position was 68 and that for the lying position 63. These 
records, which appear with but few exceptions in the curves for the 
latter part of the fast, are of special interest, as they show the change in 
the pulse-rate due to change in position. This subject will be con- 
sidered in a later section. 

On May 13-14 (figure 17) a number of observations were made in the 
afternoon, one at 2 h 30 m p. m., while the subject was talking in a lively 
manner to an assembly of medical men. That the after-effect of this 
stimulus continued for some time is shown by the curve which follows. 
The fall in the pulse-rate, due to a change in position from sitting while 
writing to lying down upon the couch, is likewise shown, as between 
6 h 20 m p. m. and 7 p. m. the subject was sitting up and writing and just 
afterward lay down upon the couch for a respiration experiment. 

On the first day of realimentation, the curve (figure 18) shows very- 
great fluctuations in the pulse-rate. These are in part due to the inges- 
tion of food and in part to the pain and distress incidental to the colic 
resulting from the taking of a large quantity of acid material into the 
stomach and intestinal tract. As a matter of fact, the highest observa- 



112 A STUDY OF PROLONGED FASTING. 

tions on this day were obtained at a time when there was a reasonably 
small amount of muscular activity. In a series of observations from 
2 h 45 m p.m. until 4 p.m., which were made while the subject was sitting 
quietly eating an orange or drinking grape juice, a value of 112 was 
found. Even an hour after eating, when the subject had colic and was 
in much distress, the pulse-rate was considerably lower than during 
the period of eating, while the average value obtained when the subject 
was lying on the couch during a respiration experiment in the early 
morning was about 59. Evidently the process of eating or drinking, 
immediately following a prolonged period of inanition, increased the 
pulse-rate very greatly. On May 15-16, the second day of this period, 
the sharp rise in the pulse-rate incidental to taking food was likewise 
noted at 9 h 40 m a. m. and again at ll b 46 m a. m. On May 17-18 the 
values obtained from 6 a. m. to 9 a. m. were unusually high, this being 
due in part to the fact that the subject was extremely excited and after 
the experimental period was over broke out into abusive language. 
There was undoubtedly a great increase in the psychic disturbance. 

COMPARISON OF PULSE RECORDS OBTAINED IN EXPERIMENTS WITH THE 
BED CALORIMETER AND THE RESPIRATION APPARATUS. 

While an examination of the general trend of the pulse curves shows 
admirably the tendency for the amplitude during the night to fall to 
a lower level, a comparison of the average values obtained under varying 
conditions can best be made in tabular form. Accordingly, in table 6 
the average values are given for observations made when the subject 
was lying in the bed calorimeter and also the average of the records 
obtained when the pulse-rate had reached its lowest level during the 
calorimeter period. The values for the experiments with the respira- 
tion apparatus are likewise given, including those made in the morn- 
ing, in the evening, when the subject was sitting quietly and also 
when writing. Furthermore, for purposes of comparison the pulse-rate 
records taken during the blood-pressure tests are included for both 
positions of sitting and lying. A number of important comparisons can 
thus be made. 

During his stay in the bed calorimeter the subject was probably asleep 
for the greater part of the time — at least on many nights. On every 
night he had periods of wakefulness, which at times may have been of 
considerable length. Consequently not all of the values obtained in 
the bed-calorimeter experiments can be taken as actually obtained 
during sleep, but by examining the curves for these experiments it is 
relatively easy to select a value which probably represents the average 
minimum pulse-rate for this subject during sleep. These values are 
given in column b in table 6. Both the average night pulse-rate and 
the average minimum pulse-rate have a distinct tendency to decrease, as 
the fast progresses, until about the twenty-second fasting day. From 



PULSE-RATE. 



113 



that time until the end of the fast the pulse records usually rise, so that 
at the end of the observations the average values are 3 or 4 beats higher 
than they were at the minimum point. 

Table 6. — Average pulse-rate of subject L. at different times of the day and with, varying activity. 













During blooc 


-pressure tests 










(about l h 30 m p.m.). 


Bed calorimeter 
















(usually 10 p.m. to 


Respira- 


Increase 






Decrease 


Date. 


Day of 
fast. 


8 a.m.). 


tion ap- 


with 
subject 






with 
ing subject 








paratus 


Sitting. 


T,y 










(subject 


awake 






lying 






Average. 


(subject 
asleep). 


awake). 1 


(c-b). 






(b— f). 






A 


B 


C 


D 


E 




F G 


1912. 


















Apr. 10-11. . 




82 
76 
78 
70 
68 


76 
70 
73 
64 
64 


272 
2 73 
2 72 
2 73 
74 


-4 
3 

-1 

9 

10 








11-12 








12-13 




13-14 . . 




14-15 


1st. . 


15-16 


2d... 


66 


63 


73 


10 










16-17 


3d... 


62 


60 


70 


10 










17-18 


4th.. 


65 


58 


68 


10 










18-19 


5th.. 


63 


59 


67 


8 










19-20 


6th.. 


60 


57 


64 


7 










20-21 


7th.. 


59 


56 


64 


8 










21-22 


8th.. 


61 


58 


65 


7 










22-23 


9th.. 


59 


67 


63 


6 










23-24 


10th. 


57 


55 


63 


8 


. , 








24-25 


11th.. 


67 


54 


61 


7 










25-26 


12th.. 


58 


56 


61 


5 










26-27 


13th.. 


66 


64 


59 


5 










27-28 


14th.. 


53 


51 


58 


7 










28-29 


15th.. 


53 


51 


67 


6 










29-30 


16th.. 


53 


62 


58 


6 




i 


H 




Apr. 30-Mayl.. 


17th.. 


52 


49 


57 


8 


68 


i 


>3 5 


May 1-2 


18th.. 


52 


51 


56 


5 








2-3 


19th.. 


52 


50 


67 


7 


69 


< 


>2 7 


3-4 


20th.. 


52 


51 


58 


7 


68 


i 


\2 6 


4-5 


21st . . 


54 


61 


69 


8 




t 


»6 


5-6 


22d. . . 


63 


61 


59 


8 


67 


t 


►8 9 


6-7 


23d. . . 


56 


53 


58 


5 


71 


e 


»3 8 


7-8 


24th.. 


55 


53 


59 


6 


76 


c 


»1 15 


8-9 


25th.. 


55 


53 


60 


7 


66 


e 


6 


9-10 


26th.. 


66 


54 


61 


7 


68 


e 


4 4 


10-11 


27th.. 


57 


55 


62 


7 


71 


c 


2 9 


11-12 


28th.. 


59 


67 


61 


4 


72 


€ 


2 10 


12-13 


29th.. 


58 


55 


63 


8 


73 


I 


7 6 


13-14 


30th . . 


58 


55 


59 


4 


69 


€ 


3 6 


14-15 


31st. . . 


57 


54 


60 


6 


3 83 


3 7 


3 10 


15-16* 




68 
64 
90 


66 
60 

84 


«72 
'84 


12 



76? 
99 


7 
8 


3 3? 

9 10 

8 


16-17 




17-18 













x The respiration experiments in the morning were usually made between 8 h 30 m and 9 b SO™. 

2 During the respiration experiments in the morning on April 11, 12, 13, and 14 the subject 
was without breakfast. 

'The subject had broken his fast by means of fruit juices during the morning. 

4 During the night of May 15-16 the subject lay on the couch in the calorimeter laboratory. 

'During the morning respiration experiments on May 17 and 18 the subject was without 
breakfast. 



114 



A STUDY OF PROLONGED FASTING. 



Table 6. — Average pvhe-rate of subject L. at different times of the day and with varying 

activity — Continued. 



Date. 


Day of 
fast. 


Respiration apparatus. 


Sitting. 1 


Lying (usually 7 to 
7 h 45 m p.m.). 


Period. 


Average. 
H 


Average. 
I 


Increase 

over 

lying 

in the 

morning 

(i-c). 

J 


1912. 

Apr. 15-16 

16-17 

17-18 

18-19 

19-20 

20-21 

21-22 

22-23 

23-24 

24-25 

25-26 

26-27 

27-28 

28-29 

29-30 

Apr. 30-May 1 . . 
May 1-2 

2-3 

3-4 

4-5 

5-6 

6-7 

7-8 

8-9 

9-10 

10-11 

11-12 

12-13 


2d 

3d 


4 h 00 m p.m. 


to 4 h 35 m p.m, 


82 
80 

62 
69 

60 
68 

68? 

69 

65 

69 

75 










4th 




5th.... 
6th 


4 10 p.m. 


4 43 p.m.* . . . 


7th 




8th 




9th.... 
10th.... 
11th 


3 52 p.m. 
3 58 p.m. 


4 28 p.m, . , 
4 57 p.m 


12th 

13th.... 
14th 


3 13 p.m. 
12 14 p.m. 


4 11 p.m 

12 48 p.m 


62 
*59 
59 
61 
59 
61 
62 
60 

57 
63 
60 
63 
63 
66 
66 
66 
67 


1 

1 
4 
1 
4 
6 
3 

-2 
4 
2 
4 
3 
5 
4 
5 
4 


15th.... 
16th 


3 23 p.m. 


3 56 p.m.*. . . 


17th.... 
18th 


9 31 a.m. 


10 04 a.m.* 


19th 




20th 

21st. . . . 


9 35 a.m. 


10 10 a.m.*.... 


22d 




23d. 
24th 


3 43 p.m. 


4 14 p.m.*.... 


25th 




26th 




27th 




28th 




29th 




13-14 


30th 


6 32 p.m. 


7 02 p.m.* 


71 


12 



1 Periods indicated by an asterisk (*) were obtained with the subject sitting, writing. 
The average pulse-rate for a period 3 h 16™ p.m. to 3 h 51 m p.m. on this day with the subject 
lying on the couch was 61 per minute. 

Any important deductions from average values for the night are out 
of the question on account of the irregularity in the number of the pulse 
records during the night and the impossibility of recording accurately 
the time when the subject slept and when he woke. If, however, we 
compare the values for the average pulse-rate with those for the aver- 
age minimum pulse-rate, we find that about the middle of the fast 
the difference is only 1 or 2 beats. The greatest variation between 
these two series of averages is on the fourth day of fasting, when the 
average during the night was 65 and the average minimum value was 
58. For further purposes of comparison, it is obviously more logical 
to use the average minimum values. 



PULSE-RATE. 115 

The pulse-rate during the respiration experiments is recorded with 
great frequency and regularity. Those obtained in the morning respi- 
ration experiments during the first period with food, i. e., in the period 
preceding the fasting period, were invariably lower than the records 
obtained during the night experiments with the bed calorimeter, with 
the single exception of the fourth record, when the average pulse-rate 
for the night had a minimum of 64 and the average obtained in the 
respiration experiment was 73. Unquestionably the pulse-rate obtained 
during the night was influenced to a considerable extent by the food 
taken the day before, especially for the evening meal. In this con- 
nection it is of special interest to note that on the last night of this 
period the food taken in the evening meal was of such a character that 
the effect would be less prolonged. From this time on, the pulse-rate 
in the experiment with the respiration apparatus was invariably higher 
than the minimum pulse record obtained during the night. The differ- 
ence in the early part of the fast was 10 beats, but then decreased until 
about the middle of the fast, when it fell as low as 5 beats. In the 
latter part of the fast it showed a tendency to increase again, although 
on the twenty-eighth and thirtieth days the difference was but 4 beats. 

This difference in the pulse records is of great significance, indicating 
clearly an increased heart action, an increased muscle tonus, and, 
according to all previous experience in this laboratory, an increased 
metabolism. The question as to the influence of sleep on the pulse-rate 
and the metabolism has received considerable attention in this labora- 
tory for a number of years and we find ourselves quite out of harmony 
with many European writers who maintain that sleep per se has no 
influence upon the pulse-rate and the metabolism. A subsequent 
examination of the records of the metabolism for this fasting subject 
bears out definitely our contention that there is a great difference in the 
pulse-rate and the metabolism in the waking over the sleeping condition. 

INFLUENCE OF BODY POSITION. 

In the latter part of the fast, the pulse-rate was recorded during the 
blood-pressure tests for both the lying and sitting positions. While the 
comparisons between the calorimeter experiments and the morning 
respiration experiments considered only the values secured when the 
subject was lying asleep or lying awake, these records obtained at noon 
indicate the change in the pulse-rate due to the change in the position 
from sitting to lying. The decrease in the pulse-rate is considerable, 
varying from 4 beats on the twenty-sixth day to 15 beats on the twenty- 
fourth day. A number of pulse records for the sitting position were 
also obtained with the subject in two morning and nine afternoon 
respiration experiments. In the morning experiments and also in four 
of the afternoon experiments, the subject was writing. On the after- 
noon of the second day of the fast, when the subject was sitting quietly, 



116 



A STUDY OF PROLONGED FASTING. 



the pulse-rate was as high as 82; the lowest observation was 60, which 
was recorded on the twelfth day of the fast. 1 

INFLUENCE OF THE WORK OF WRITING. 

As the subject spent a considerable portion of the day sitting up 
writing, an effort was made to study the pulse-rate during a period of 
writing. Hence on a number of days the subject was connected with 
the respiration apparatus, the metabolism was studied, and records of 
the pulse-rate were taken. On two of the days these tests were carried 
out in the morning, immediately after the regular series of respiration 
experiments. On these two days — the seventeenth and twentieth days 
of the fast — the combined influence of this position and occupation was 
to increase the pulse-rate over the records obtained in the lying position, 
this increase being 12 beats on the seventeenth day and 7 beats on the 
twentieth day. Four experiments of this character were also made in 
the afternoon, but the pulse-rate showed little increase over records 
obtained in similar afternoon experiments when the subject sat quietly 
without writing. 

INFLUENCE OF BREATHING AN OXYGEN-RICH ATMOSPHERE. 

In an earlier series of observations of the pulse-rate in which normal 
subjects breathed an oxygen-rich atmosphere, there was a distinct 
tendency for the pulse-rate to decrease. Since it was of interest to note 
if this tendency to a decrease in the pulse-rate would be greater in 
prolonged fasting than under normal conditions, respiration experiments 
were made on the twenty-eighth, twenty-ninth, and thirtieth days of 
the fast, in which the subject breathed an oxygen-rich atmosphere and 
pulse records were simultaneously obtained. These experiments imme- 
diately followed the morning respiration experiments for those days in 
which the subject breathed air with a normal content, the average 
pulse records for the latter experiments being 61, 63, and 59 respectively. 
The pulse records in the experiments with the oxygen-rich atmosphere 
were as shown herewith. 



Date. 


Time of day. 


Pulse-rate. 


May 12.... 
13.... 
14.... 


9 h 04 m a.m.-9 h 48 m a.m. 
9 06 a.m.-9 52 a.m. 
9 16 a.m.-9 49 a.m. 


62 
61 

58 



From these results it appears that the pulse-rate in the high-oxygen 
experiment on the first day increased by 1 beat, on the second day 
decreased by 2 beats, and on the third day decreased by 1 beat. In 
former experiments 2 with normal individuals, the average decrease in 

x The designation of the days for part of the data in table 6 is not in strict accordance with our 
method of ending the experimental day with the end of the morning respiration experiment, 
but the same numbering is used throughout for comparison purposes. 

2 Benedict and Higgins, Am. Journ. Physiol., 1911, 28, p. 1. 



PULSE-RATE. 117 

the pulse-rate with the oxygen-rich atmosphere was not far from 5 to 
6 beats, but with this fasting subject the inhalation of an oxygen-rich 
atmosphere apparently produced no change in the pulse-rate. 

DIURNAL RHYTHM. 

Beginning with the twelfth day of the fast, respiration experiments 
were made each evening, just before the subject entered the bed calori- 
meter. Comparing the records of the pulse-rate for these experiments 
with those obtained in the morning experiments, it will be seen that 
the pulse-rate in the evening was almost invariably higher than in the 
morning. While the difference in the middle of the fast is very slight, it 
tends to increase as the fast progresses, until on the thirtieth day there is 
a difference of 12 beats. On the twenty-first day the evening rate was 
2 beats lower than that observed in the morning, the lowest pulse 
records for the evening experiments being found on this day. 

IRRITABILITY OF THE HEART. 

In all of the comparisons of the pulse records it is difficult to find 
any very definite indications of the so-called " irritable heart," espe- 
cially emphasized by the Berlin investigators in their study of Cetti. 
It is true that the change from lying asleep to lying awake resulted in an 
increase in the pulse-rate, the difference in the fasting period being not 
far from 7 to 8 beats, with a maximum of 10 and a minimum of 4 beats. 
It is further true that the change from a sitting to a lying position, as 
noted at the time of the blood-pressure tests in the latter part of the 
fast, tended to decrease the pulse-rate from 4 to 15 beats per minute, 
averaging not far from 7 to 8 beats. Likewise, the pulse-rate in the 
evening respiration experiments averaged a few beats more than in the 
morning respiration experiments. But no great indication of an irrita- 
bility of the heart was noted with any of these minor changes in position 
and activity. 

On the food days, however, there was a great increase in the pulse- 
rate at the time the food was taken. Furthermore, the records obtained 
at the time of the blood-pressure observations show that the pulse-rate 
did not quickly return to the minimum after changing from a sitting 
to a lying position for the pulse-rate in the lying position is usually a 
few beats higher than the value obtained in the morning respiration 
experiments. Since the value for the lying position was found almost 
immediately after the test, it is hardly possible that the body had time 
to adjust itself to the position, but the tendency to reach the minimum 
found in the morning is worthy of note. Even with the slight activity 
due to the blood-pressure tests and changing the position, the pulse- 
rate is not so high as the pulse-rate obtained in the evening respiration 
experiments, showing that the prevailing diurnal variation was greater 
than that obtained with slight activity earlier in the day. 



118 A STUDY OF PROLONGED FASTING. 

A point of considerable importance is the fact that there was a 
distinct tendency for the pulse-rate to reach a minimum between the 
fourteenth and twenty-second days of the fast, that is, during the third 
week. Consequently, all of the pulse records, including the average 
minimum records inside the bed calorimeter, the average for the calori- 
meter experiments, and the average for the morning respiration exper- 
iments, show a tendency to increase in the fourth week. The same 
tendency may be noted in the pulse records obtained at various times 
throughout the day. It is clear, therefore, that during the latter part 
of the fast the heart of the subject was in a somewhat more irritable 
condition than during the third week. 

The observations on the pulse-rate in this fasting experiment have 
great significance when compared with the simultaneous measurements 
of the metabolism, as is done in subsequent sections of this publication. 
The records have therefore been presented in extenso, as they show 
strikingly that the pulse-rate may legitimately be used as an index of 
the metabolism. The total metabolism was measured only during the 
times when the subject was on the respiration apparatus or inside the 
bed calorimeter. For the remainder of the day, especially during those 
times when the subject was most active, the metabolism was not meas- 
ured, so that the probable metabolism at these times must be estimated 
in so far as possible from the records of the pulse-rate. We therefore 
have little evidence of the effect of muscular activity upon the heart, 
except as shown by the few pulse records taken following slight activity. 
These do not indicate a distinctly irritable heart. 

The muscular activity of the subject was probably greater at the time 
of the dynamometer tests than during any other observations. The 
relatively few records made before and after these tests show a distinct 
rise in the pulse-rate incidental to the dynamometer test, but also a very 
rapid return. Unfortunately they were not taken with sufficient 
regularity for us to note positively any indication of the increased or 
decreased irritability of the heart as the fast continued. It was the 
intention to study in this experiment the influence of light muscular 
activity upon the heart beat and the metabolism, but here again the 
unwillingness of the subject to engage in muscular activity of any kind 
prevented valuable observations originally planned for. 



BLOOD PRESSURE. 

As it seemed desirable to supplement the observations of the pulse- 
rate throughout the fast with observations of the blood pressure, the 
determinations were made by Mr. H. L. Higgins. In accordance 
with the advice of Dr. E. P. Cathcart, who had previously experimented 
with Beauts, the values were secured for both the lying and sitting 
positions, as it was quite possible that later in the fast it might be 
advantageous to keep the subject in bed. The records were made 
almost invariably shortly after noon and immediately alter the subject 
had taken a glass of water. Employing the auscultatory method with 
the Erlanger sphygmomanometer, both the systolic and diastolic pres- 
sures were obtained. 

In former fasts emphasis has been chiefly laid upon the determina- 
tions of the systolic pressure. Luciani, in using the sphygmomanom- 
eter of von Basch with an aneroid manometer, found that the varia- 
tions in the blood pressure at the radial artery had a tendency to 
decrease as the fast continued, the decrease being from 220 mm. on 
the first day of fasting to 120 mm. on the twenty-sixth day. On the 
last four days of the fast there was a slight tendency for the pressure to 
increase. Luciani also states that there were daily fluctuations from 
morning to evening, which frequently, but not always, corresponded 
to changes in the temperature and the pulse-beat. 

Cathcart, 1 using the C. J. Martin modification of the Riva-Rocci 
apparatus, found with Beaute" a continual fall in the maximum pressure 
during the 14 days of the fast, his values being 108 for the first day and 
for the subsequent days 96, 98, 98, 92, 94, 88, 90, 88, 92, 88, 94, 90, 
and 88. These observations were always taken in the evening. Cath- 
cart's determinations were controlled by Charteris, 2 who used the same 
instrument and presumably made the records at about the same time. 
The values differ but little from those reported by Cathcart, showing 
an excellent agreement of blood-pressure determinations by two obser- 
vers. Charteris concludes that the pulse-wave became shorter and 
weaker, but remained regular in rhythm, and that the arterial pressure 
gradually sank so that at the end of the fast the fall amounted to 
almost 25 per cent of the normal reading. He noted a rapid recovery 
after the fast, the pressure being again practically normal after the first 
week of food. 

In a 30-day fast reported by Penny, 3 in which he was himself the 
subject, the author states that the blood pressure was taken by a 
Martin's modification of the Riva-Rocci sphygmomanometer and fell 

Cathcart, Biochem. Zeitschr., 1907, 6, p. 109. 
2 Charteris, Lancet, 1907, 173, p. 686. 
'Penny, British Medical Journal, 1909, p. 1414. 

119 



120 A STUDY OF PROLONGED FASTING. 

steadily during the fast from 1 10 to 90 mm. The daily observations are 
not recorded. 

No direct blood-pressure measurements were made by the Berlin 
investigators on their fasting subjects, although from an examination 
of the pulse curves made with a Marey sphygmograph at the radial 
artery, Senator and Mueller 1 concluded that there was a noticeable 
decrease in the arterial tension which produced not only a dicrotism 
but likewise a decrease in the elasticity. This was more noticeable 
with a subject who fasted 10 days than with another who only fasted 
6 days, notwithstanding the fact that on the last fasting day the pulse- 
rate was so weak that they could not secure a suitable curve. 

The observations made of the blood pressure in the fasting experi- 
ment with our subject L. are given graphically in figure 19, the curves 
representing the systolic, diastolic, and pulse pressures for both posi- 
tions of lying and sitting. With these are compared curves showing 
the average pulse-rate secured in the bed-calorimeter experiments 
throughout the night and also in the morning respiration experiments. 
During the latter part of the fast, the pulse-rate was likewise secured 
at the time the blood pressure was taken, in both positions of lying and 
sitting. These values are given in table 6 (pages 113 and 114). 

As is usually the case, the systolic blood pressure when the subject 
was lying down was invariably somewhat higher than when the subject 
was in a sitting position, with a general tendency for the difference 
between the two to become greater as the fast progressed. On the last 
day of the fast, however, the difference was not much greater than at 
the first of the fast. The curves for the diastolic pressure also show 
higher values for the lying position, although the difference is not so 
great as with the systolic pressure, for up to about the fifteenth day 
the two curves are approximately the same. 

The systolic pressure for the lying position falls quite rapidly through 
the first half of the fast, fluctuating considerably in the last half above 
or below the average value of 100 mm. of mercury. The records ranged 
from 134 mm. on April 16 to 94 mm. on April 30. An even lower value 
was obtained on May 16, 26 hours after taking food, namely, 92 mm. 
Two days later, however, it had increased to 124 mm. 

The curve for the systolic pressure for the sitting position is nearly 
parallel to that of the systolic pressure for the lying position, although 
in the latter part of the fast the values were considerably lower and the 
fluctuations were not so great. The range in values was from 123 mm. 
on April 16 to 83 mm. on May 8. During the latter part of the fast 
it averaged not far from 90 mm. 

The diastolic pressure for the lying position shows a marked fall from 
the third to the fourth day. Subsequently there is a gradual fall to the 
middle of the experiment, with a distinct tendency in the latter part of 

^ehmann, Mueller, Munk, Senator, and Zuntz, Archiv f. path. Anat. u. Physiol, u. f. klin. 
Med., 1893, 131, Supp. p. 101. 



BLOOD PRESSURE. 



121 



APRIL MAY 

II 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 I 2 3 4 5 6 7 8 9 10 1 1 12 13 14 15 16 17 18 
DAYS Or DAYS OF FASTING DAYS OF 

F00D I 2 3 4-5 6 7 8 9 10 1 1 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 F00D 




Fig. 19. — Chart showing blood pressure, pulse pressure, and pulse-rate of subject L. 



122 A STUDY OF PROLONGED FASTING. 

the fast for the general course of the curve to rise. The highest record 
was 100 mm. on April 16 and the lowest was 73 mm. on May 6. Twenty- 
six hours after food was first taken the diastolic pressure in this position 
fell to 74, but it rose on the third day to 102 mm. 

A similar course is followed by the curve for the diastolic pressure in 
the sitting position, although the fall from the third to the fourth day is 
not so pronounced. The slight tendency to rise in the last part of the 
fast is also shown in this curve. The highest record was 90 mm. on 
April 16 and the lowest was 70 mm., recorded on April 30, May 1, and 
May 2. 

The general trend of all four curves for the systolic and diastolic 
pressures in the two positions is a distinct decrease in blood pressure 
during the first 15 days of the fast, followed either by an average con- 
stant value or a slight tendency for the pressure to rise in the last part 
of the fast. 

With these observations of the systolic and diastolic pressure, it was 
possible to obtain the pulse pressure, this being the difference between 
the maximum and minimum blood pressures. According to some 
writers, the pulse pressure is of more significance than the blood-pres- 
sure observations themselves. An examination of the curves shows 
that the pulse pressure in the lying position was, except on one day, 
higher than the pulse pressure for the sitting position. The curves are 
by no means parallel, however, as there are great differences in the 
levels. The pulse pressure for the lying position decreases until the 
fifteenth or sixteenth day of the fast, and then fluctuates considerably, 
but averages approximately 23 mm. The pulse pressure for the sitting 
position shows a much more marked similarity to the trend of the blood- 
pressure curves, i. e., a falling pressure until about the fifteenth day of 
the fast, followed by a period of approximately constant values, and 
finally a distinct tendency to increased pressure towards the end of 
the fast. 

In considering the blood-pressure records, it is of interest to compare 
them with the average pulse-rate curves obtained in the bed calorimeter 
experiment during the night and on the respiration apparatus in the 
morning. These curves are given in the lower part of figure 19, and 
show a general parallelism with each other. Curves are also given 
showing the pulse-rate records for both positions obtained in the latter 
part of the fast at the time the blood pressure was taken. A fact of 
special interest in connection with these curves is that the average 
pulse-rate in the latter part of the fast has a tendency to rise during the 
calorimeter experiment and also in the morning respiration experiment. 
During the same period the blood-pressure curves remain essentially 
constant, showing only a slight tendency to rise, the pulse pressure for 
the sitting position alone having a tendency to follow more closely the 
average pulse record. From this it may be concluded that the pulse- 



BLOOD PRESSURE. 123 

rate, which is considered in this laboratory as an index to the total 
metabolism, is not so indicative of the actual work of the heart as of 
the general metabolic tonus of the whole body. 

The distinct decrease in blood pressure as the fast progressed may be 
due to one or both of two causes. The first is the decrease in the con- 
tractibility of the heart muscles as the fast continued and the second 
is the decrease in the general tone of the peripheral vessels. Luciani 
points out that theoretically the heart muscles should decrease more 
than any other muscles of the body, as they are continually at work, and 
that such a decrease has been found with fasting subjects. As will be 
seen from Dr. GoodalPs report, 1 there was evidence of a distinct and 
regular decrease in the size of the heart during the fast. It is true that 
the decrease in the size of the heart observed by Dr. Bianchi, on Succi 
in his Florence fast, was not so great as that found by Dr. Goodall with 
L. ; nevertheless the general conclusion in both series of observations is 
the same, namely, that there is a considerable diminution in the size 
of the heart as the fast progresses. 

Somewhat in opposition to the belief that the size of the heart is the 
determining factor in the decrease in the blood pressure is the fact that 
on the third day after the fast the systolic and diastolic pressures had 
both returned to their normal level; the complete regeneration of 
the heart muscles in this time is difficult to conceive. A disturbing 
factor entered into the observations of the last day as the subject was 
laboring under intense psychical excitement. 

^ee page 65. 



THE BLOOD. 

By J. E. Ash, M. D., 
Department of Pathology, Harvard Medical School. 

CORRELATION OF LITERATURE. 

The blood has frequently been studied during inanition, but an 
exhaustive search through the literature brought to light only a few 
records of systematic examinations covering so long a period of fasting 
in man as the case that forms the basis of this report. Fasts of man 
conducted under scientific supervision and including blood examina- 
tions are limited in number, though the work on animals has been 
rather prolific. Before presenting the findings in Levanzin's blood, 
there is given a correlation of abstracts from this literature, of interest 
not only from an historical standpoint, but in demonstrating the diver- 
sity of results obtained by the various observers. 

Among the earliest references are those to the work of Valentin 1 in 
1838, who concluded that there was no alteration in the relation of 
blood-weight to body-weight as a result of exhaustive starvation, and 
that of Bidder and Schmidt, 2 in 1852, though these latter do not report 
anything more specific than an increase in the solid constituents in 
the blood of a starving cat. With reference still to the blood as a 
whole, London, 3 much later studying a series of 8 rabbits from which 
both food and drink were withheld, found a loss in total quantity, 
proportional, though, to loss of body-weight. Pashutin 4 concludes, 
as the result of the work of Heidenhain, Panum, and Voit, that the 
blood is not impoverished by fasting; on the contrary, in certain periods, 
the organism is plethoric. He holds it as remarkable that the number 
of erythrocytes increase — probably, however, only because of the rapid 
decrease in plasma. This latter fact is demonstrated markedly in the 
dogs observed by W. Muller and Buntzen, 6 though in none of their 
animals did a loss of more than 15 per cent in body-weight occur. 
Luciani 6 holds that, aside from water-content, the blood exhibits a 
resistance similar to the nervous system and that the apparent fluctua- 
tions, in corpuscular content at least, depend chiefly on the amount of 
water consumed. Chossat, however, quoted by Pashutin, 4 considered 

Valentin, Repel, f. Anat. u. Physiol., 1838, 3, p. 156. 

2 Bidder u. Schmidt, Die Verdauungsf ahrte in der Stoffwechsel, Mitau u. Leipzig, 1852, p. 328. 

3 London, Note sur la question du changement de la quantite generate et de l'alcalinite du sang 
dans le jeune absolu. Arch, des Sciences Biol., 1895-96,4, p. 516. (Abstract by Miihlmann. See 
footnote 4, p. 125.) 

4 Pashutin, Pathological Physiology, Inanition. 1902, 2, pt. i, p. 81 (Russian). 

6 Miiller and Buntzen, Transfusion and plethora. Christiania, 1875. 

•Luciani, Fisiologia del digiuno, Firenze, 1889. Authorized translation by M. O. Fraenkel. 
Das Hungern, Studien u. Experimente am Menschen, Hamburg u. Leipzig, 1890. 

124 



THE BLOOD. 125 

that, next to the fat, the blood suffered the greatest loss, amounting even 
to 75 per cent of its former weight. This view is not tenable in the light 
of practically all other work and was evidently the result of faulty 
technique or observations. Pashutin 1 quotes Valentin as noting 
striking general changes during hibernation — in part, that the blood 
putrefies from 2 to 4 times more slowly, that the arterial blood is not so 
bright a red, and that the venous blood is not so dark as normally, due 
to disturbance of oxygen interchange. 

ERYTHROCYTES. 

Considering more specifically the blood elements and beginning with 
the erythrocytes, we find that as early as 1843 Schultz 2 studied starving 
animals and found these cells atrophic, attributing the death of the 
animals to the inability of the shrunken cells to bind oxygen. Jones, 3 
but a few years later (1856), observed that the corpuscles in dogs' 
blood appeared to have undergone "partial decomposition." Others 
since then have noted these striking alterations in shape and size of the 
corpuscles, among them Manassein, 4 Andral-Gavarret, 4 Laptschinski, 4 
and especially Kagen, 6 who studied dogs and rabbits. He found little 
change in the first days, but as the fast progressed the red cells became 
smaller and crenated ones appeared more frequently, until at the end 
many "star forms" were seen and microcytes predominated. Liu- 
boumdrow 6 also found variations in the character of the red cells — 
macrocytes, microcytes, and nucleated cells being common, especially 
the large form, which reached 20 to 30 per cent of the total red count. 

In the roundabout way, we get from Pashutin 7 an abstract from 
Wratsch, 1881, p. 78, quoting from foreign journals (not specified) a 
reference to Dr. Tanner's blood after his 40-day fast in 1880. This was 
a public exhibition, but well controlled and was absolute for the first 15 
days. The plasma and white cells presented nothing unusual. The red 
cells, however, were somewhat smaller than normal, being ^Vinr m °h m 
diameter instead of 40 1 00 to -ginnr inch. 

Curtis 8 made systematic observations of Griscom's blood during 
his 45-day fast in 1880. This constitutes the longest period with blood 
examinations of which any record could be found. 

'Pashutin, Pathological Physiology, Inanition, 1902, 2, pt. i, p. 81 (Russian). 

2 Schultz, Beitr. x. phys. u. path. Chem. von Simon, 1884 (quoted by Muhlinann. See footnote 
4, this page.) 

3 Jones, Smithsonian Cont. to Knowledge, 1856. 

4 Muhlmann, Russisch Literatur iiber die Pathologic des Hungerns. Centralblatt f. allgem. 
Path., 1899, 10, p. 160. 

'Kagen, Blood and blood pressure in starving organisms. Dissert. , St. Petersburg, 1884, Russian. 
From the Laboratory for General and Experimental Path., Prof. V. Pashutin, St. Petersburg. 

6 Liuboumdrow, Changes in the blood and organs in starvation. 71 Dissert., 1893, Russian. 
From the Path. Anat. Laboratory, Prof. W. Winogradow, St. Petersburg. 

7 Pashutin, Pathological Physiology, Inanition, 1902, 2, footnote p. 605 (Russian). 

'Curtis, A study of blood during a prolonged fast. Proc. Am. Ass. Adv. Science, 1881, 30, 
pp. 95-105. 



126 A STUDY OF PROLONGED FASTING. 

Curtis describes the morphology of the erythrocytes as follows : 

The first examination, made just after Griscom's last meal, showed the cells 
in abundance, of bright color, regular, smooth of outline, solid in appearance, 
and of usual size — y-rVir inch. 

On the third day they were paler and apparently not so firm. 

Fourth day: The change had progressed. There were two sorts of cells to be 
seen, one pale and large, the other deeper in color and contracted. Some of 
the former were almost invisible, appeared soft and sticky, enveloping objects 
encountered in flow. Their shape was altered to a round rim with abrupt 
descent to a flat floor. They averaged -ohm inch. The other sort were 
deeper in color, less transparent than normal, and covered with nodules like 
blunt cones (evidently crenated). The cells had lost their usual concavity, and 
seemed as though acted upon by an astringent, being much smaller than 
normal — tjVtt inch. 

Fifth day: The soft pale cells had disappeared, the smaller variety seemed 
larger and nodular. Irregularities in shape were first noticed, some cells being 
elongated, others lemon or club shaped, and still others had pointed ends. 

Sixth to ninth days: The large soft form appeared and persisted in greater 
or less numbers. Later small colored bodies like red corpuscles appeared, 
measuring T¥ Vv inch. All the erythrocytes at this time were small. The 
extremely small ones continued to increase in number and diminish in size. 

Sixteenth day: Corpuscle-like bodies observed as small as yuVxr inch. Those 
of nfon to tuVtt inch diameter were like normal red cells. Others were nodu- 
lated or of a chestnut-burr appearance. 

Thirty-sixth day: " Saw an erythrocyte undergo direct division. From this 
day on, the red cells changed for the worse." They became pale, ragged and 
shrivelled. At this time the subject showed signs of weakened circulation — 
vertigo, numbness of hands and feet. 

Thirty-eighth day: He fainted on rising from bed. 

Thirty-ninth day: There was scarcely a normal corpuscle to be seen. 

Fortieth day: After an excursion of 2\ hours on the lake, there was a remark- 
able change in the blood picture. The ragged, pale, and broken corpuscles all 
disappeared and all the erythrocytes became smooth in outline and bright in 
color. They seemed quite normal, except that they were smaller, averaging 
7T Vtf inch. After this, they again retrograded, became soft, pale, and sticky, 
but never so bad as just before the lake excursion. Certain minute granules 
were seen in this blood, granules which, in the author's experience, exist in all 
other persons, except one, whose blood was examined. They were small red 
points Tu^irTr inch in diameter and highly refractile. They are found also in 
lymph and cow's milk. They existed in the blood in great numbers at first, 
decreasing till after the eighth day; then disappearing until the twenty-fourth 
day, when a few pale ones appeared. They then increased in number, but 
only returned to their normal abundance after the fast was broken. (These 
were apparently the platelets that he was observing.) 

This report is given in detail to illustrate not only the painstaking 
care with which the observations were made, but somewhat the com- 
parative crudity of the methods employed at that time. As will be 
seen, it is only in the earlier reports that the appearance of macrocytes, 
microcytes, crenated and distorted cells are recorded, and it has 
occurred to the writer that with the improvement of blood technique 
the occasion for their presence was eliminated. 



THE BLOOD. 127 

In 1887 Senator 1 reported finding a number of microcytes on the 
13th day of a " Schlafsucht" in which a 54-year-old woman lay for about 
7 weeks receiving as nourishment only a small amount of milk and wine. 
In Succi's 2 blood, late in the 40-day fast made as a public exhibition 
in London in 1890, numbers of imperfect blood disks were observed. 
Charteris, 3 on the other hand, could find no alteration in shape, size, 
or staining qualities of the erythrocytes in the blood of his human 
subject during the fast of 14 days in 1907, except that a few nucleated 
ones did appear during the last 4 days. Two examinations were made 
on Gayer's blood by Dr. Wile. 4 His was a public fast of 30 days, 
undertaken largely for advertising purposes. He drank water ad lib., 
and though he was fairly well guarded, there was opportunity for his 
obtaining food secretly. However, at the beginning of the fast his 
weight was 210 pounds and at the end 174f pounds, a loss of over 35 
pounds, which proves rather conclusively that the experiment was con- 
ducted in good faith. On the eighteenth day, the red cells numbered 
5,192,000 and on the thirtieth, 5,776,000, a slight rise. The absence 
of anisocytosis or any degeneration of cells is mentioned specifically in 
the report. The subject refused to allow further examinations, particu- 
larly after breaking the fast, as desired by Dr. Wile in order to determine 
a normal picture. In none of the other reports are the characteristics 
of the individual red cell noted, so it is most likely that alterations of 
importance did not occur. 

Continuing the consideration of the effects of inanition on the numer- 
ical estimation of the erythrocytes, the results are found to be at rather 
wide variance. 

Senator 1 found no significant variation during the long period of 
almost complete inanition already mentioned above. He attributed 
little value to the small number of counts he made, but concluded 
there was probably a slight diminution. During Cetti's 10-day and 
Breithaupt's 6-day fasts, studied by Senator, Lehmann, et ah? an 
increase was noted, amounting, in the former subject, to a million. 
Cetti's normal count of 5,720,000 was above the average. (Table 7.) 

Dup6ri6 6 also claims that a considerable increase in number occurs. 
From the study of Succi's blood during a 30-day fast supervised by 
Luciani, 7 the latter concludes that the variations noted in numbers are 

Senator, Ueber einen Fall von sog. Schlafsucht mit Inanition. Charit6-Annalen, 1887, 12, p.316 

2 The fasting man. Brit. Med. Journ., 1890, 1, p. 1444. 

3 Charterid, Record of changes observed in the blood count and in the opsonic power of a man 
undergoing a prolonged fast. Lancet, 1907, 2, p. 685. 

*Gayer's fast: A private communication from Dr. Wile, of New York City. 

5 Lehmann, Mueller, Munk, Senator, and Zuntz, Untersuchungen an zwei hungernden Men- 
schen. Archiv f. path. Anat. u. Physiol., Virchow's, 1893, 131, supphft., p. 1. 

6 Duperie, Sur les variations physiologiques dans l'etat normal des globules du sang. Paris, 
1878. Cited by Rollett in Hermann's Handbuch d. Physiolog., 4, (1). 

7 Luciani, Fisiologia del digiuno, Firenze, 1889. Authorized translation by M. O. Fraenkel. 
Das Hungern, Studien u. Experimente am Menschen, Hamburg u. Leipzig, 1890. 



128 



A STUDY OF PROLONGED FASTING. 



only relative, depending on the concentration or dilution of blood from 
alteration in water-content. While on the twenty-seventh day the 
greatest loss was noted, there followed on the twenty-ninth day a rise 
that brought the number to the level of the first day. Andreesen, 1 
Malassez, 1 and Lepine 1 found that while in the beginnings of the fasting 
periods there would be an increase in number, in the later days a 
decrease occurred. 



Table 7. — Red-cell counts on Ceiti and Breithaupt. 



Day. 


Cetti. 


Breithaupt. 


3d fast 


6,720,000 

5^285,000 

6,830,000 

6,660,000 
6,730,000 


4,953,200 
6,184,000 

'4,801,000 

'4,820,000 
4,812,000 


4th fast 


6th fast 


9th fast 


2d diet 


2 weeks later 



'Before and after first meal on the sixth day. 

Clinical records furnish us with the two following reports of relevant 
interest. The first is of one of Landouzy's patients studied by 
Malassez. 2 

"A boy of 18 lived 3 months and 20 days with a stricture of the esophagus, 
the result of swallowing H 2 S0 4 . He obtained practically no nourishment, as 
he vomited food administered by tube. Ten days before death, he began to 
take a small quantity of milk and meat. The red cells numbered 3,600,000 
20 days before death, and a week before the end 2,600,000, a decided loss 
from normal. Two days before the boy died, the count had risen to 3,200,000. 
A transfusion was performed immediately after this examination, followed in 
20 minutes by another count which showed a rise to 3,500,000. The next day 
they had returned to 3,200,000." 

This is a striking loss, and while it is not possible to rule out a toxic 
influence in this case, the almost complete inanition was no doubt the 
prominent factor. 

"The second case (reported by Brouardel) 3 was a man of 48 years, who lived 
4 months 12 days after an experience similar to that just quoted. But one 
blood examination was made, and that 2 days before death, when the erythro- 
cytes were 4,849,000 and the leucocytes 7,852." 

While to the present writer these would be considered as practically 
normal counts, the author of the report concludes that they demonstrate 
a concentration of the blood. 

'Miihlmann, Russiche Literature iiber die Pathologie des Hungerns. Centralblatt f. allgem. 
Path., 1899, 10, p. 160. 

2 Malassez, Bull, et mem. de la soc. med. des hopitaux de Paris, 1874, 11, p. 124. 
"Brouardel, Union Med., 1876, ser. 3, 22, p. 408. 



THE BLOOD. 



129 



Von Noorden 1 has found the corpuscular content normal in five cases 
of gastric ulcer with emaciation. He remarks that in spite of the anaemic 
appearance presented by patients suffering from various conditions caus- 
ing malnutrition, their blood is usually normal. This of course does not 
hold in those cases where the cause of the malnutrition has a direct 

Table 8a. — Estimations of red cells during Griscom'sfast (Curtis). 



Day. 


Estimation 
of red cells. 


Remarks. 


4th 


4,320,000 




5th 


4,485,000 




6th 


2,370,000 




8th 


4,860,000 




10th 


3,260,000 




11th 


4,720,000 




12th 


3,790,000 




13th 


4,480,000 




14th 


4,210,000 




15th 


2,800,000 




18th 


5,790,000 




19th 


6,770,000 




20th 


6,500,000 


Flatulence. Patient felt quite ill ; 


21st 


5,600,000 


took enema, causing stool. 


22d 


2,100,000 




23d 


5,460,000 




24th 


5,420,000 




25th 


3,920,000 


These figures are not entirely des- 


26th 


4,160,000 


titute of symmetry. 


27th 


2,540,000 




28th 


3,130,000 




29th 


3,180,000 


Counts on 6 intervals of 6 days. 


30th 


3,180,000 


10 " " 4 " 


31st 


3,360,000 


15 " " 5 " 


32d 


4,420,000 


22 " " 7 '* 


33d 


3,600,000 


27 " " 5 " 


34th 


3,900,000 


37 " " 10 " 


35th 


3,700,000 


40 " " 3 " 


36th 


3,810,000 


44 "4 " 


37th 


3,520,000 


Pointing to the opinion held of a 


38th 


4,080,000 


certain limited duration of life 


39th 


4,200,000 


of red blood corpuscles. 


40th 


3,200,000 




41st 


3,390,000 




42d 


3,590,000 




43d 


3,490,000 




44th 


3,150,000 




45th 


5,390,000 





influence on the blood, as in infections. He discusses this practical phase 
of the subject and gives many references to observations of the effects 
on the various properties and constituents of the blood of "clinical" 
inanition. 



x Von Noorden, Metabolism and practical medicine, Anglo-American issue, Chicago, 1907, 2, 
p. 28. 



130 A STUDY OF PROLONGED FASTING. 

Returning to Curtis's 1 article, we find the following protocol (table 8 a) 
of the 38 numerical estimations made of the red cells during the 45-day 
fast of Griscom, with the former's comments. 

"The subject was at his worst physically and mentally between the twenty- 
seventh and fortieth days, and during this period the counts were consistently 
low. On the fortieth day he took the excursion on the lake, which was appar- 
ently the cause of the drop of 1,000,000 from the count of the preceding day. 
It will also be noted that for the few days before a decided fall in number there 
was usually a rise. The corpuscles on the days of these low counts always 
appeared healthier than at other times. On the last day Mr. G. drank no 
water and the high count of that day may have been due to concentration." 

Kagen, 2 in 1884, claimed that the ordinary methods of determining 
the cell content of blood were open to so many sources of error that the 
results were not dependable. He limited his observations, therefore, 
to the direct estimation of the solid constituents, the specific gravity 
(by pyknometer), and haemoglobin content (Malassez's hsemochromom- 
eter) . He examined 6 dogs and found in the early days an increase in all 
three factors, attributing the changes to concentration through water 
loss. The amount of solid constituents, he claims, can equal even at the 
end of the fast that present under normal conditions. Liuboumdrow 3 
noted, as an average of observations on 17 dogs, a slight increase in 
erythrocytes till the loss of body- weight amounted to 10 to 15 per cent, 
then a steady decrease till death, the diminution amounting to as high 
as 32 percent on the twenty-eighth day. Nasse 4 also found an increase 
in number in a dog after 11 days of complete fasting. As proof that 
this was due to variation in water-content, he states that he obtained a 
reaction in the opposite direction when the animal was again allowed 
water. 

Pol6taew 5 studied 8 dogs that received neither food nor water, dying 
after loss of 50 per cent in body-weight. These all showed an increase 
in red cells until late in the fasts, after a loss of 30 percent body- weight, 
when there was a gradual decrease till death. Pol6taew is not satisfied 
with the explanation of this finding simply on the grounds of concen- 
tration, for he found an increase also in the dogs that were allowed 
water. He holds that while there may be interference with blood 
formation, there is also less destruction for bile formation. Tauszk, 6 

Curtis, Physiology of autonutrition ; A study of blood during a prolonged fast. Am. Ass. 
Adv. Science, 1880, 30, pp. 95-105. 

2 Kagen, Blood and blood pressure in starving organisms. Dissert. St. Petersburg, 1884, 
Russian. From the Laboratory for General and Experimental Path., Prof. V. Pashutin, St. 
Petersburg. 

3 Liuboumdrow, Changes in the blood and organs in starvation. 71 Dissert., 1893, Russian. 
From the Path. Anat. Laboratory, Prof. W. Winogradow, St. Petersburg. 

4 de Martigny u. Nasse, Ueber den Einfluss der Nahrung auf daa Blut. Marburg u. Leipsic, 
1850. 

6 Poletaew, The morphologic composition of the blood in complete and incomplete starvation 
in dogs. Dissert. 97, 1894, St. Petersburg (Russian). From the Laboratory of Path. Anat., 
Prof. Uskow, Riv. internaz. d'ig., Roma, '95, 6, p. 129, and Arch. d. sc. Biol. St. Petersburg, 
1893. 2, p. 794. 

6 Tauszk, Jahrsb. tiber d. Fortschr. der Thier-Chemie, 1894, 24, p. 147; abstracted from Orvoia 
hetilap, Budapest, 1894, p. 512. Also Haematologische Untersuchungen am hungernden mens- 
chen. Wien. klin. Rundschau, 1896, 10, p. 306. 



THE BLOOD. 



131 



in a study of Succi's blood during his 30-day fast in 1894, found, after a 
short interval of decrease, a moderate increase in the red cells. (See 
table 8 b.) The form of the cells remained normal to the end. 

Daiber 1 in 1896 drew his conclusions as to the effect of inanition on 
the blood from his findings in Succi's urine during a 20-day fast. There 
was an increase in urobilin and earthy alkaline phosphates, both of 
which were to be accounted for by assuming an enormous destruction 
of erythrocytes, though he does concede that the phosphates might 
have come from tissue destruction elsewhere. As proof of the adapta- 
bility of the blood to altered conditions, he presents the decrease in 
urobilin and disappearance of phosphate sediment noted after the 
fifth day, showing an acquired resistance to the previously destructive 
influence of fasting. The urobilin was distinctly demonstrable through- 

Table 8 b. — Succi's red-cell counts. 







Ratio 


Day. 


Red cells. 


of red to 
white cells. 


Third 


5,246,000 


1 :545 


Eighth 


4,840,000 


1 :584 


Thirteenth. . . 


4,932,000 


1 :684 


Seventeenth . . 


5,136,000 


1 :744 


Twenty-first. . 


5,160,000 


1 :938 


Twenty-fifth. . 


5,268,000 


1 : 1097 


Thirtieth 


5,472,000 


1 : 1302 



out the fast, though greatly reduced, but the phosphate sediment was 
replaced by one of urates. The plasma remained intact, as no transu- 
dation of its constituents, particularly albumen, through the kidneys 
could be demonstrated. Daiber concludes that the conditions during 
inanition must resemble those present in continued fevers in which there 
is usually red-cell destruction sufficient to give rise to a demonstrable 
anaemia. In these cases urobilin is present in the urine in distinctive 
amounts. 

In the dog which died on the twenty-fifth day, after a loss of 52 per 
cent in body-weight, Hayem 2 reports an increase till the eighteenth 
day from 4,200,000 to 5,500,000. There was then a slight decrease, 
though at the end the erythrocytes numbered 4,800,000, still above the 
original count. The hsematoblasts decreased continually during the 
fast. Reyne 3 found a progressive increase in the number in a dog 
dying on the twenty-fifth day of starvation. Charteris, 4 on the other 

'Daiber, Beitrag zur Kenntniss des Stoffwechsels beim Hungern. Schweitzer Wochenschr. f . 
Chem. u. Pharm., Zurich, 1896, 34, p. 395. 

2 Hayem, Lecons sur les modifications du sang. Paris, 1882, p. 382. 

3 Reyne, quoted by E. Bardier in his article on Inanition, in Dictionnaire de Physiologie, 
Charles Richet, 9, p. 99. 

4 Charteris, Record of changes observed in the blood count and in the opsonic power of a man 
undergoing a prolonged fast. Lancet, 1907, 2, p. 685. 



132 A STUDY OF PROLONGED FASTING. 

hand, in the case of his already mentioned, could find no suggestive 
variation, though there was some daily fluctuation. 

Gordon 1 studying the blood of Martin, a medical student who under- 
went a 9-day fast with the uniform daily water consumption of 24 
ounces, could find practically no variation in red-cell count, except that 
on the sixth day of refeeding it was about 1,000,000 below the normal. 
At the end of the first week of Succi's fourth fast, one of 40 days con- 
ducted in London, 2 the red cells numbered 6,500,000, an increase of 
1,000,000 over the average normal individual's count. It may be more 
or less in this particular case, as the normal count is not given. A. R. 
Diefendorf , however, found a slight diminution during and a relatively 
rapid rise immediately following each of the two fasts of a man of 7 and 
4 days interrupted by a feeding period of 19 days, which formed the 
basis of Benedict's 3 report. 

Three counts were made on the blood of Dr. Penny, 4 who fasted for 
30 days, in 1909, drinking only distilled water. They demonstrated a 
moderate increase till the twentieth day and a loss of 1,000,000 during 
the remaining 10 days. Here again no normal count was obtained. 
The results for the three counts were for the twelfth day, 6,600,000; 
twentieth day, 7,000,000; thirtieth day, 6,000,000. Ronsse and van 
Wilder 5 hold there will always be a slow increase in erythrocytes if 
water as well as food is withheld. 

Though the conditions are not altogether analogous, it is interesting 
to note that in hibernating animals there is a decided decrease in erythro- 
cytes, as is reported by Ranke. 6 

HAEMOGLOBIN. 

Senator, using the v. Fleischl method, noted a moderate increase in 
his woman subject 7 and in Breithaupt; 8 a loss, however, of about 20 
per cent in Cetti 8 in 9 days. (Table 9.) 

Liuboumdrow 9 (with Malassez's method) found a slight increase in 
the blood of dogs until a loss of 10 to 15 per cent of body-weight had 
occurred, when a decrease was recorded that progressed until the 
animals died. 

Benedict 3 reports a slight loss during, with a rise after, the fasts, 
corresponding to the fluctuations of the erythrocytes. The v. Fleischl 
and Tallqvist methods were used. 

Gordon, A prolonged fast. Montreal Med. Journ., 1907, 36, p. 482. 

2 The fasting man. Brit. Med. Journ., 1890, 1, p. 1444. 

»Benedict, Carnegie Inst. Wash. Pub. 77, 1907, p. 322. 

*Penny, Notes on a thirty-day fast. Brit. Med. Journ., 1909, 1, p. 1414. 

'Ronsse et van Wilder, Variations du nombre des globule3 rouges et du taux de l'hemoglobine 
au cours de l'inanition chez le lapin. Arch, intern, de Pharm. et Ther., 1903, 11, p. 301. 

6 Ranke, Grundziige der Physiolog., 3d ed. t p. 380. 

"Senator, Ueber einen Fall von sog. Schlafsucht mit Inanition. Charite-Annalen, 1887, 12, p. 316. 

8 Lehmann, Mueller, Munk, Senator, and Zuntz, Untersuchungen an zwei hungernden Men- 
schen. Archiv f. path. Anat. u. Physiol., Virchow's, 1893, 131, supphft., p. 1. 

'Liuboumdrow, Changes in the blood and organs in starvation. 71 Dissert., 1893, Russian. 
From the Path. Anat. Laboratory, Prof. W. Winogradow, St. Petersburg. 



THE BLOOD. 



133 



Martin's 1 haemoglobin was 90 per cent the first day, 95 per cent the 
fourth, and 90 per cent again on the last day, the ninth, dropping to 80 
per cent 6 days after resuming food. 

Penny 2 showed an increase of 8 per cent during his 30-day fast, going 
from 104 to 112 per cent. 

Table 9. — Haemoglobin estimations on Cetti and Breithaupt. 



Day. 



Before fast 

Second day 

Third day 

Fourth day 

Sixth day (before first meal) 

(2 hours after first meal) 

Ninth day 

Second day of diet 



Cetti. 



Per cent. 
115-118 



110 
85-90 



Breithaupt. 



Per ant. 
107 
114 
114 
110 
130 
116 

114 



In Luciani's 3 report is to be seen a variation synchronous with that 
of the erythrocytes, except on the eleventh and thirteenth days, when 
the percentage failed to rise with them. There was a small increase on 
the third and twenty-first days and it was lowest on the thirteenth and 
twenty-sixth days. The highest estimation was 90 per cent, the lowest 
72 per cent (v. Fleischl apparatus). He concludes that there is an 
actual loss of haemoglobin. Quoting his studies with Bufalini, carried 
out in Sienna in 1882, on a dog that lived 53 days without food, he states 
there was a rapid rise during the first 6 days. (Modified Bizzozero's 
method used). This he explains as being due not only to concentration 
from loss of water, but to the more rapid consumption of plasma than 
corpuscles. The initial rise was followed by a gradual decrease con- 
tinuous until the last 12 days of the fast, during which the percentage 
was constant. 

Gayer' s 4 percentage rose from 80 on the eighteenth day to 100 on the 
thirtieth, when his fast was broken. 

Charteris, 6 reporting the 14-day fast, notes a drop from 110 per cent 
to 96 per cent, after remaining unaffected for the first few days. The 
loss was not recovered until several days after breaking fast. 

Subbotin, 6 using Preyer's method (spectroscope), found a decrease 
of haemoglobin in a case fed on nitrogen-free diet. By the twenty-sixth 
day it had fallen from 13.80 per cent to 1 1.65 per cent and on the thirty- 

'Gordon, A prolonged fast. Montreal Med. Journ., 1907, 36, p. 482. 

2 Penny, Notes on a thirty-day fast. Brit. Med. Journ., 1909, 1, p. 1414. 

3 Luciani, Fisiologia del digiuno, Firenze, 1889. Authorized translation by M. O. Fraenkel. 
Das Hungern, Studien u. Experimente am Menschen, Hamburg u. Leipzig, 1890. 

4 Gayer's fast: A private communication from Dr. Wile, of New York City. 

6 Charteris, Record of changes observed in the blood count and in the opsonic power of a man 
undergoing a prolonged fast. Lancet, 1907, 2, p. 685. 

6 Subbotin, Zeitschr. f. Biol., 1871,7, p. 187. 



134 



A STUDY OF PROLONGED FASTING. 



eighth it was 9.52 per cent. In starving rabbits, however, there was 
an increase due (he concludes) to decrease in water-content of the blood. 
In a dog that starved for 38 days there was very little variation, from 
13.80 at the beginning to 13.33 per cent at the end. 

Though the inanition was only partial, the experiment of v. Hoss- 
lin 1 is suggestive and his conclusions are interesting. He observed 
two growing dogs, one of which, (a), weighing 3.2 kilograms, was given 
only one-third the nourishment that (6), weighing 3.1 kilograms, 
received. Table 10 presents the results: 

Table 10. — Results of v. Hdsslin's observations. 



Dog. 


66th day. 


124th day. 


18 months. 


Kilos. 


Percent- 
age of 
haemo- 
globin. 


Kilos. 


Percent- 
age of 
haemo- 
globin. 


Erythro- 
cytes. 


Kilos. 


Percent- 
age of 
haemo- 
globin. 


Erythro- 
cytes. 




6.5 

11.6 


11.2 
10.2 


8.5 
23.4 


16.0 
14.9 


7,970,000 
6,820,000 


9.5 

30.3 


15.5 
17.6 


7,300,000 
8,300,000 



The difference in haemoglobin content of the two dogs, in spite of 
extreme emaciation in dog (a), is still within physiological limits, 
proving how independent of the amount of nourishment the haemoglobin 
really is. He claims a greater influence on the blood constituents 
through the nature than the amount of nourishment, long-continued 
impoverishment in albumen, e. g., causing a decrease in both haemo- 
globin and red cells. We do wrong, he claims, to consider as a result 
of the malnutrition the apparent or real anaemia seen in poorly nourished 
individuals. Either it is only (1) apparent, the haemoglobin and red- 
cell content remaining high while the patient appears anaemic from 
contraction of peripheral vessels in an effort to compensate for the 
lessened thermogenesis, or it may be (2) actual anaemia, the result, 
however, of the condition that is responsible for the malnutrition, e. g., 
long-continued fever, cancerous ulcers, repeated haemorrhages, intes- 
tinal parasites, etc. The amount of nutrition, however, has a great 
influence on blood formation, so that anaemia, from whatever cause, 
will clear up much more readily under good than poor nourishment. 
The effect of uncomplicated inanition, therefore, is to reduce the total 
quantity of blood as it does the muscle and organ volume generally, 
rather than disturb the individual constituents. On refeeding, this 
total quantity is restored quickly, so that there appears an anaemia, in 
spite of the better nourishment — a relative condition, however, due to 

V. Hosslin, Ueber den Einfluss ungeniigender Ernahrung auf die Beschaffenheit des Blutes. 
Gesellschaft f. Morphologie u. Physiol., Munich, 1890, p. 119. 



THE BLOOD. 135 

more rapid restoration of fluid than haemoglobin and cellular elements 
and apparent till the normal relation is established. 

Gallerani 1 found that the mean resistance of the haemoglobin of 
dogs and frogs to solutions of NaClof various percentages was increased 
during fast; the resistance to the high percentage solutions decreased, 
while to the low it increased. The former was due, he claimed, to the 
absence of newly formed haemoglobin, which is more resistant to the 
stronger solution and the latter to the absence of very old or much used 
haemoglobin, which is less resistant to the weak solution. 

Hermann 2 in discussing the subject, says that the changes found 
with the ordinary methods of examination can well be due to concen- 
tration, because the water is the most variable of the blood constituents. 
Results are conclusive, therefore, only when they deal less with the 
haemoglobin content than with the relation of haemoglobin to the total 
quantity of solid constituents. 

Groll's 3 work was carried out on this line with rabbits, cats, and one 
dog. Their fasts were absolute, no water being allowed. They existed 
under these conditions for from 1 to 22 days. He estimated the haemo- 
globin quotient by dividing the percentage of haemoglobin, as deter- 
mined by the v. Fleischl apparatus, by the percentage of solid con- 
stituents. The latter was obtained by heating measured quantities of 
blood to 110° C. till the weight remained constant. Withfew exceptions 
the color quotient thus found was increased. The simple haemoglobin 
per cent showed a rise in all the animals, being, however, only a relative 
increase, due to concentration of the blood. The diminution in the 
total solids was due to the greater susceptibility to destruction of the 
other solid constituents. During the period of restitution, there was 
a diminution in the haemoglobin. This, again, was largely a relative 
change, occurring as a result of dilution from increased water intake, 
though no doubt the haemoglobin is slower in regenerating as well as 
being more resistant to destruction. Groll concludes that the haemo- 
globin is more stable in starving conditions than any of the other solid 
constituents of the blood. 

It would seem from these data that the red cells and haemoglobin 
are particularly resistant, though in the long fasts there is no doubt a 
slight loss of both elements. The consensus of opinion appears to be 
that concentration of the blood through water loss is responsible for 
the increase found during the fast and that dilution from more rapid 
return of water than the other elements accounts for the decreases 
found immediately after the fasting. 

1 Gallerani, Resistenz des Haemoglobins im Hunger. Jahrsb. u. die Fortschr. der Thier- 
Chemie, 1894, 24, p. 120. (Abstracted from Ann. di chim. Farmacol., 1892, 15, p. 3.) 

2 Hermann, Untersuchungen iiber den Hsemoglobin-Gehalt des Blutes bei vollstandiger Inani- 
tion. Dissert., Kdnigsberg i. Pr., 1887. 

s Groll, Untersuchungen iiber den Haemoglobin Gehalt des Blutes bei vollstandiger Inanition. 
Dissert., Konigsberg i. Pr., 1887. 



136 



A STUDY OF PROLONGED FASTING. 



LEUCOCYTES. 

More attention has been paid to the white cells than to any of the 
other blood constituents and reports are more at variance as to just what 
does happen to them during states of inanition. Almost every possible 
change, especially numerical, has been observed at one time or another. 

Morphologic alterations are recorded by Luciani, 1 who noted an early 
decrease in size, so that by the fifth day all the leucocytes were smaller 
than the red cells. They recovered their normal size by the ninth 
day, however. 

Charteris, 2 on the other hand, mentions specifically that he observed 
no alteration in size. 

Manassein 3 reports the presence in the leucocytes of fasting rabbits 
of refractile bodies that are not affected by acetic acid. 

Kallmark 4 observed, in his rabbits, rarefaction in the basophiles, 
with agglutination and peripheral arrangement of the granules. 

Curtis, 5 in 1880, observed peculiar bodies, resembling leucocytes but 
larger, consisting of spherules too small to measure. These cells meas- 
ured ttsW inch in diameter and exhibited amoeboid movement. When 
these very indefinite bodies were most abundant, the granules were 
absent from the leucocytes. Curtis does not speculate as to whether 
these were altered white corpuscles or a foreign cell entering the blood 
from the tissues. 

Hayem 6 concludes that there is no essential change in the leucocytes 
during starvation, at least in dogs. 

Considered numerically both as to total and differential estimations, 
the following results are reported from studies of fasts in man : 

Table 11. — Cetti' 8 and Breithaupt' 8 white-ceU count. 
Cetti fasted 11 days, Breithaupt 6. 



Day. 


Cetti. 


Breithaupt. 


Before fast 


Normal. 
4,800 

4,200 

12,300 
7,950 


6,500 

6,870 
7,000 


Fourth day 


Sixth day 


Ninth day 


Broke fast 


Second day of diet .... 



luciani, Fisiologia del digiuno, Firenze, 1889. Authorized translation by M. O. Fraenkel. 
Das Hungern, Studien u. Experimente am Menschen, Hamburg u. Leipzig, 1890. 

2 Charteris, Record of changes observed in the blood count and in the opsonic power of a man 
undergoing a prolonged fast. Lancet, 1907, 2, p. 685. 

3 Muhlmann, Russiche Literature iiber die Pathologie des Hungerns. Centralblatt f. allgem. 
Path., 1899, 10, p. 160. 

4 Kallmark, Zur Kenntniss des Verhaltens der weissen Blutkorperchen bei Inanition. Folia 
Hsemat., 1911, 11, pt. 1, p. 411. 

*Curtis, Physiology of autonutrition : A study of blood during a prolonged fast. Am. Ass. Adv. 
Science, 1880, 30, pp. 95-105. 

6 Hayem, Lecons sur les modifications du sang. Paris, 1882, p. 382. 



THE BLOOD. 



137 



With both Cetti and Breithaupt 1 a moderate decrease was observed 
during, with a considerable rise for the first few days following, the 
fasting period. (See table 11.) 

Senator 2 concludes that there is a lively new formation of leucocytes 
on refeeding. 

Luciani 3 records a marked diminution in the early period of Succi's 
30-day fast, dropping from 14,536, the count on the first day, to 861 on 
the seventh. The count then rose to 1,550, where it remained until the 
twenty-ninth day with slight fluctuations due to concentration and 
dilution of blood. He attributes the marked diminution to the diges- 
tive action of trypsin, which evidently enters the blood as such during 
the cessation of intestinal digestion. He bases this theory on the work 
of Albertoni, who by intravenous injection of trypsin got almost a 
complete disappearance of leucocytes. The trypsin apparently has 
no effect on the erythrocytes. It is quite possible, also, that there may 
exist in the early days of the hunger period some special destructive 
condition, evidenced also by the loss of haemoglobin. Two other factors 
at work, he argues, are, first, the disappearance of the lymphocytes, they 
no longer being required to alter the assimilated products of digestion 
in the blood plasma (after the work of Schaeffer, Hofmeister and 
Zawarykins); secondly, the leucocytes may have lost their "Wander- 
lust." There would, then, not only be a failure of "outwandering" 
from the blood, but of "inwandering" from the tissues as well, and the 
latter would exert the greater influence on the number. The white 
cells practically disappeared from Succi's blood during his fourth fast, 
till late, when small and ill-formed corpuscles were found. 

Tauszk 4 notes a decrease in total count during one of Succi's 30-day 
fasts. As will be seen in table 12 a this was due to loss of mononu- 

Table 12 a. — Succi's total and differential white-cell counts. 



Day. 


Total 
white 
cells. 


Poly- 
morphs. 


Mono- 
cytes. 


Eosino- 
phils. 


Third 

Eighth 

Thirteenth. . . 
Seventeenth. . 
Twenty-first . 
Twenty-fifth. 
Thirtieth 


9,600 
8,300 
7,200 
6,900 
5,500 
4,800 
4,200 


p. ct. 
64.1 

68!5 

79!2 


p. ct. 
33.1 

27 A 

16.6 


p. ct. 
2.7 

3.9 

4.7 



1 Lehmann, Mueller, Munk, Senator, and Zuntz, Untersuchungen an zwei hungernden Men- 
schen. Archiv f. path. Anat. u. Physiol., Virchow's, 1893, 131, supphft., p. 1. 

'Senator, Bericht tiber die Ergebnisse des auf Cetti ausgefuhrten Hungersversuchs. Berlin 
klin. Wochenschr., 1887, 24, p. 427. 

3 Luciani, Fisiologia del digiuno, Firenze, 1889. Authorized translation by M. O. Fraenkel. 
Das Hungern, Studien u. Experimente am Menschen, Hamburg u. Leipzig, 1890. 

4 Tauszk, Jahrsb. iiber d. Fortschr. der Thier-Chemie, 1894, 24, p. 147, abstracted from Orvosi 
hetilap, Budapest, 1894, p. 512. Also Hsematologische Untersuchungen am hungernden Men- 
schen. Wien. klin. Rundschau, 1896, 10, p. 306. 



138 



A STUDY OF PROLONGED FASTING. 



clear cells, including lymphocytes. The eosinophiles and polymorphs 
were increased. Neubert 1 found the opposite changes, that is, an 
increase in mononuclear cells and decrease of the eosinophiles and 
polymorpho-nuclears. His studies were made on cases of carcinoma 
and pulmonary tuberculosis, so that the inanition was not simple. 

There was little change in the white-cell content of Martin's 2 blood 
during his 9-day fast, except that on the second and ninth days they 
rose to 10,000. As will be seen in table 12 6, there was a very slight 
progressive loss in the polymorphs, while the lymphocytes were 
increased somewhat on the sixth and ninth days. There was no 
differential count made to specify the total rise noted on the second 
day. 

Table 12 b. — Martin's differential white-cell counts. 



Day. 


Poly- 
morphs. 


Small 
mono- 
nuclears. 


Large 
mono- 
nuclears. 


Transi- 
tionals. 


Baso- 
phils. 


Eosino- 
philes. 


Sixth day 


p.ct. 
68 
60 
65 
59 
58 


p. ct. 
22 
30 
18 
32 
35 


p.ct. 

6 

7 
23 

9 

6 


p. ct. 
4 

2 
1 


p.ct. 

3 
1 


p. ct. 
3 


Ninth day 

Sixth day after .... 



The results of the two examinations of Gayer's 3 blood are given in 
table 13. The striking points in this case are the very low total count 
on both occasions, the increase in the small and the drop in the large 
lymphocytes at the end of the fast. No explanation of these changes is 
offered by Wile, who made the observations. 

Table 13. — Gayer's total and differential white-cell counts. 



Day. 


Total 
white 
cells. 


Poly- 
morphs. 


Large 
lymph- 
ocytes. 


Small 
lymph- 
ocytes. 


Eosin- 
ophiles. 


Baso- 
phils. 


Thirtieth 


2,600 
2,800 


p.ct. 
54 
51 


p.ct. 

12.0 

5.4 


p.ct. 
30.4 
42.0 


p. ct. 
1.6 
1.0 


p.ct. 

2.0 

.6 



In Professor Benedict's subject, whose blood was studied by Dr. 
Diefendorf, 4 there was evidently normally a high white-cell count. 
The results are given in table 14. 

During the first fast of 7 days there was a progressive diminution till 
the last day, when there was a slight rise. After an interval of 19 days 
there was a second fast of 4 days. During this period a gradual rise 

x Neubert, Ein Beitrag zur Blutuntersuchung, speciell Phthisis pulm. und carcinom, Dorpat, 
1889. 
2 Gordon, A prolonged fast. Montreal Med. Journ., 1907, 36. p. 482. 
3 Gayer's fast: A private communication from Dr. Wile, of New York City. 
^Benedict, Carnegie Inst. Wash. Pub. 77, 1907, p. 322. 



THE BLOOD. 



139 



amounting to about 2,000 was noted. The polymorphs averaged high 
during both fasts, but at no time could the number be considered 
distinctly pathological. The small lymphocytes averaged low. The 
large lymphocytes were high during the last 2 days of the first period 
and throughout the last. The eosinophiles were low and the baso- 
phils high in both fasts. 

Table 14. — S. A. B.'a total and differential white-cell counts. 





Total, 
white 
cells. 


Poly- 
morphs. 


Small 
lymph- 
ocytes. 


Large 
lymph- 
ocytes. 


Eosino- 
philes. 


Mast 
cells. 


1905. 
Prior to fast: 

Mar. 3 , , . 
Fasting: 

Mar. 6 1 

7.,. 
8,.. 
9 
10 

11 
Diet: 

Mar. 12 

13... 
14 

15 , . 
20 ... 

Apr. 3 

7 , 
Fasting: 

Apr. 8 

9 
10.... 
11 
Diet: 

Apr. 12 ... . 
25 ... 

26 

May 5 


23,000 

15,750 
13,000 
12,000 
10,000 
10,000 
10,000 
11,250 

13,250 
13,750 
14,500 
15,000 
11,750 
13,000 
13,250 

13.250 
13,250 
13,750 
14,500 

15,000 
5,500 
8,100 


p. ct. 

63 
76 
73 
76 

77 
64 
75 

66 

66.5 

69 

78 
58.5 
54 
63 

69 
73 
76 

68.5 
62 
50 
49 


p. ct. 

27.0 
15.5 
17.5 
18.0 
10.5 
25.0 
10.3 

17.5 
20.5 
17.0 
14.5 
22.5 
25.0 
14.5 

15.0 
18.3 
21.0 

27.3 
33.5 
35.5 
45.0 


p. ct. 

7.3 
7.3 
8.0 
5.1 

11.5 
9.1 

12.7 

13.4 
14.4 
10.8 
6.9 
15.7 
17.0 
13.9 

12.9 
11.8 
15.4 

9.1 
3.1 

10.9 
5.8 


p. ct. 

2.0 
0.5 
1.6 
0.4 
1.2 
1.4 
0.45 

1.6 

0.0 

1.65 

1.8 

2.4 

1.8 

1.0 

1.3 
1.6 
0.0 

0.8 
1.6 
1.8 
0.25 


p. ct. 

0.6 

.4 
.5 
.6 
.4 
.6 

.2 

.2 

.2 

.4 

1.0 

1.0 

1.0 

1.0 
1.0 

.4 

.4 

.4 

1.0 



x The first fasting day was March 4-5. 

Charteris, 1 in 1907, found a moderate leucocytosis, reaching 14,000 
on the sixth day from 5,300, the count before the fast. He noted 
further a gradual increase of the eosinophiles to 7 per cent, a condition 
that had never been noted before in human blood during inanition. 
His subject went 14 days without food, receiving a constant quantity of 
water, 1 liter per day. 

Penny's 2 blood showed also a slight leucocytosis, with a return to 
normal at the end. Only three counts were made, these with the 

1 Charteris, Record of changes observed in the blood count and in the opsonic power of a man 
undergoing a prolonged fast. Lancet, 1907, 2, p. 685. 

2 Penny, Notes on a thirty-day fast. Brit. Med. Journ., 1909, 1, p. 1414. 



140 



A STUDY OF PROLONGED FASTING. 



results shown in table 15. The noteworthy features of the differential 
counts are the high polymorph percentage, the very marked falling-off 
of the lymphocytes, and the increase in the large mononuclears. 

Reyne 1 could demonstrate no influence on the leucocytes in his dog 
that fasted for 25 days. 

Howe and Hawke 2 studied two men during 7-day fasting periods 
with uniform water allowance. There was an increase in the poly- 
morphs at the beginning, followed by a decrease to below normal by 
the end of the fast. The small lymphocytes presented the reverse 
picture, while the large lymphocytes increased during the early days. 
One subject showed a moderate increase in eosinophiles. The blood 
of both men returned to normal after several days of diet. 

Table 15. — Penny's total and differential white-cell counts. 



Day. 


Total. 


Poly- 
morphs. 


Large 
mono- 
nuclears. 


Lymph- 
ocytes. 


Eosino- 
philes. 


Twelfth 

Twentieth.. . . 
Thirtieth, . . , 


10,000 

11,000 

8,800 


p. ct. 
76 
76 
70 


p. ct. 
12 
18 
20 


p. ct. 

12.0 

6.0 

7.5 


p. ct. 
1.5 



Even in animals where conditions can be comparatively readily con- 
trolled, there has been a striking lack of harmony in the findings. 
Rabbits and dogs have been the animals of most frequent choice and 
Okintschitz's 3 report of his work on the leucocytes of the former in 
1892 is one of the earliest. He was concerned only with the differential 
variations. Normally in rabbits the eosinophiles constitute about 50 
per cent, lymphocytes 25 per cent, large round cells and polymorphs each 
12.5 per cent of the total white-cell content. Following Professor Luk- 
janow's classification of the inanition period, he divides it into four parts : 
(1) the stage of indifference; (2) that of excitation; (3) of depression; 
(4) paralysis of functions. The animals were allowed no water. He 
found a diminution of the lymphocytes and polymorphs and an increase 
in eosinophiles and large round cells. During the middle periods, the 
relative diminution was not so rapid as in the first and last periods. 
The lymphocytes and mononuclears showed their respective alterations 
in the first period, while the polymorphs and eosinophiles were not 
affected till later. The polymorphs showed the most marked diminu- 
tion and on refeeding they were increased, seeming, therefore, to be the 

1 Reyne, quoted by E. Bardier in his article on Inanition, in Dictionnaire de Physiologie, 
Charles Richet, 9, p. 99. 

2 Howe and Hawke, Fasting studies, No. IX. On the differential leucocyte count during pro- 
longed fasting. Am. Journ. Physiol., 1912, 30, p. 174. 

3 Okintschitz, Ueber die Zahlenverhaltnisse verschiedener Arten weisser Blutkorperchen bei 
vollstandiger Inanition undbei nachtraglicher Auffilterung. (Versuche an Kaninchen). Archiv f. 
exp. Path. u. Pharm., 1892-93, 31, p. 383. 



THE BLOOD. 141 

form most affected by food. The disturbance in blood picture was 
still evident, even when the animals had almost completely regained 
their body-weight. Hayem, 1 much earlier, in 1882, could see no differ- 
ence in the variations of the leucocytes in the dog studied by him during 
its 25-day fast, and those which occur under normal circumstances. 

Argaud and Billard 2 found about the same alterations in the blood 
of the two rabbits they studied, as did Okintschitz. They report a 
marked hypoleucocytosis with an inversion of the formula, there being 
present 3 mononuclears to every polymorph. The recovery in these 
animals, however, was more rapid, for in a few days the blood picture 
had resumed the normal. 

Kallmark 3 also studied rabbits during periods of complete starvation 
varying from 7 to 14 days. He noted a primary fall in lymphocytes 
and polyneutrophiles, followed by a rise until the seventh day, slow in 
the latter form, more rapid in the lymphocytes. In the longer fasts 
this rise was followed by fluctuations, none of which, however, went as 
high as the normal counts. The basophiles showed a marked rise on 
the third day of the longer fasts. On refeeding, the polymorphs showed 
a more rapid rise than the lymphocytes, duplicating the experience of 
Okintschitz. Kallmark concludes that the lymphocytes are supplied 
in greater abundance during inanition, most probably by the thymus, 
which in so doing atrophies. The primary fall and the post-inanition 
rise in the leucocytes occur, he believes, before compensation for the 
disturbance of equilibrium has been established or the organism has 
adapted itself to the altered conditions. When this has been accom- 
plished, the changes in the blood are not so much different from those 
noted under normal circumstances. 

Rieder 4 reports finding a marked hypoleucocytosis in the dogs he 
studied in 1892. 

Liuboumdrow 5 found that the leucocytes of his 15 dogs decreased 
gradually at the beginning of their fasts or until a loss of 20 per cent 
body- weight was reached. A gradual rise was then noticed, except in 
6 of them, frequently reaching normal. The lymphocytes showed a 
diminution persisting to the end, most marked early, dropping from 
15 per cent to 3 per cent or less. The monocytes reappeared to a 
certain extent, after the primary fall. The polymorphs were propor- 
tional throughout to the total count. Eosinophils appeared early in 
those animals that did not show them before fasting, and in most cases 
there was an increase of 7 to 8 times, which lasted until a loss of from 

^ayem, Lecons sur les modifications du sang. Paris, 1882, p. 382. 

2 Argaud et Billard, Inversion de la formule leucocytaire sous l'influence de l'inanition. Compt. 
rend. Soc. Biol., 1911, 70, p. 746. 

'Kallmark, Zur Kenntniss des Verhaltens der weissen Blutkorperchen bei Inanition. Folia 
Haemat., 1911, 11, pt. 1, p. 411. 

4 Rieder, Beitrage zur Kenntniss der Leukocytose u. s. w., Leipsic, 1892. 

6 Liuboumdrow, Changes in the blood and organs in starvation. 71 Dissert., 1893, Russian. 
From the Path. Anat. Laboratory, Prof. W. Winogradow, St. Petersburg. 



142 A STUDY OF PROLONGED FASTING. 

10 to 30 per cent in body-weight had occurred, when they began to 
diminish. By the end of the fast they had disappeared altogether. 

Pol^taew 1 and Reyne 2 both observed great variations in the number 
of white cells in the dogs they studied, the former concluding that there 
was evidently a diminution in all the forms until a loss of from 30 to 
40 per cent in weight and then an increase toward the end, of the 
younger elements, including lymphocytes. 

Uskow 1 interprets these results as follows: In the beginning the 
entrance of young leucocytes into the blood is retarded, as is also the 
transition of the young into ripe forms. In the later period, however, 
the lymph tissue, probably stimulated into increased activity by the 
products of degeneration, sends more cells into the blood, and further, 
there is probably a more rapid development of the young forms already 
present into ripe cells. 

Keuthe 3 noted a decrease in polymorphs and an increase in lympho- 
cytes during the first days and a reversal of this relation in the later 
days of fasting. 

Pashutin, 4 on the other hand, concludes that the fast has practically 
no effect on the leucocytes, that they show very little alteration. 

Howe and Hawke 5 observed the following changes in four dogs 
receiving only a constant quantity of water: Three of them, fasting 117, 
15, and 30 days, respectively, showed a decrease in polymorphs with 
an increase in the small lymphocytes. The basophiles, eosinophiles, 
and transitional forms showed no noteworthy changes. The fourth 
dog, fasting for 48 days, was already anaemic. His blood presented the 
reverse picture, the polymorphs increased, the lymphocytes decreased. 
An early-developing eosinophilia disappeared. Two of these four 
animals showed an increase in large lymphocytes, while the other two 
showed a fairly constant decrease in the same variety of cell. During 
later fasting periods of 15 and 30 days in these dogs, the results were 
quite different, all the forms remaining practically constant, save the 
large lymphocytes. 

The work of Mann and Gage 6 is of interest, though it is concerned with 
the effects of food rather than of starvation on the morphology and stain- 
ing properties of the leucocytes. They conclude that during digestion 
there is a marked increase in the intensity of staining in the nuclei; the 

1 Pol6taew, The morphologic composition of the blood in complete and incomplete starvation 
in dogs. Dissert. 97, 1894, St. Petersburg (Russian). From the Laboratory of Path. Anat., 
Prof. Uskow., Riv. internaz. d'ig., Roma, 1895, 6, p. 129. and Arch. d. sc. Biol., St. Petersburg, 
1893, 2, p. 794. 

2 Reyne, quoted by E. Bardier in his article on Inanition, in Dictionnaire de Physiologie, 
Charles Richet, 9, p. 99. 

3 Keuthe, Ueber die funktionelle Bedeutung der Leukocyten im Zirkulierenden Blute bei 
verschiedener Ernahrung. Deutsch. med. Wochenschr., 1907, 33, p. 588. 

4 Pashutin, Pathological Physiology, Inanition, 1902, 2, pt. i, p. 81 (Russian). 

6 Howe and Hawke, Fasting studies, No. IX. On the differential leucocyte count during pro- 
longed fasting. Am. Journ. Physiol., 1912, 30, p. 174. 

•Mann and Gage, On the changes induced in blood by feeding, etc. Lancet, Lond., 1912, 2, 
p. 1069. 



THE BLOOD. 143 

rim of cytoplasm in the lymphocytes becomes narrower; the granules in 
the leucocytes decrease both in size and number; and the entire cell may 
show a diminution in size. 

Considering again the question of hibernating animals as throwing 
some light on the changes during simple starvation, the work of Hanse- 
mann 1 is interesting. Killing the animals during their hibernating 
state and examining the various organs, no evidence of cell division 
could be found. He concludes that the physiologic cell division occurs 
as a result of the mechanical wearing out of the tissue. If this is 
eliminated, as it is under these circumstances when muscular and diges- 
tive activity and the general vital processes are practically in abeyance, 
there is no stimulus for cell division. The reduced activity, the absence 
of intestinal digestive processes or products, and the presence of per- 
verted products of parenteral digestion during fasting, no doubt would 
be the important factors in influencing the blood picture. 

Argaud and Billard 2 examined the blood in hibernating dormice and 
only a few monocytes were found after careful search, the other forms 
having apparently disappeared. 

Valentin, quoted by Pashutin, 3 had the same experience, finding 
only a few white cells. He explains their absence to the lack of lymph, 
which, he claims, introduces the leucocytes into the blood. 

Interesting, also, in view of the findings in some of the cases of inani- 
tion quoted above, are the changes noted in bone marrow by Roger and 
Josue. 4 In rabbits that were completely starved for 6 or 7 days the 
marrow showed the presence of many giant cells. Neutrophilic granu- 
lar myelocytes predominated, though there were many polymorpho- 
nuclear cells. Eosinophilic cells were rare. The fat was largely re- 
placed by a granular, albuminoid substance, not mucin. On refeeding, 
the eosinophiles were even less in evidence, but there were many very 
large giant cells and numbers of nucleated red cells, some of them 
polynuclear. Not before 24 days of feeding did the marrow return 
to its normal state. This picture is not altogether consistent with 
the blood findings reported, especially the eosinophilia described by 
Okintschitz 5 and the scarcity of polymorphs observed by Argaud and 
Billard. 2 

As already stated, the thymus has been found to atrophy during 
inanition. Kallmark 6 not only noticed diminution in the size of the 

1 Hansemann, Ueber den Einfluss des Winterschlafes auf die Zellteilung. Archiv f. Physiol., 
1898, 5 and 6, p. 262. 

2 Argaud et Billard, Inversion de la formule leucocytaire sous l'influence de l'inanition. Compt. 
rend. Soc. Biol., 1911, 70, p. 746. 

3 Pashutin, Pathological Physiology, Inanition, 1902, 2, pt. i, p. 81 (Russian). 

4 Roger et Josue, Des modifications histologiques de la raoelle osseuse dans l'inanition. Compt. 
rend. Soc. Biol., 1900, 52, p. 417. 

6 Okintschitz, Ueber die Zahlenverhaltnisse verschiedener Arten weisser Blutkorperchen bei 
vollstandiger Inanition und bei nachtraglicher Auffuterung (Versuche an Kaninchen). Archiv 
f. exp. Path. u. Pharm., 1892-93, 31, p. 383. 

"Kallmark. Zur Kenntniss des Verhaltens der wie3sen Blutkorperchen bei Inanition. Folia 
Hsemat., 1911, 11, pt. 1 p. 411. 



144 A STUDY OF PROLONGED FASTING. 

organ, but in the number of mitotic figures, an evidence of its inactivity. 
He quotes v. Friedleben as having made the same observations as early 
as 1859, Hammar in 1905, and v. Jonson in 1909. These are the 
only references that could be found mentioning the histological appear- 
ances of the haematopoietic organs during starvation. This is a neg- 
lected feature of the subject that would seem to offer a rich field for 
investigation. Curran 1 and Jolly and Levin 2 write of the general patho- 
logical changes. The latter carried out their studies on rats and describe 
particularly the changes in the lymphatic tissue, essentially, atrophy, 
particularly of the Malpighian bodies. 

PHYSICO-CHEMICAL CHANGES. 

Specific gravity. — The question of influence of food and drink and the 
abstinence from them on the density of the blood has been rather 
frequently the subject of investigation. We find that, as early as 1834, 
Thacrah 3 noted an increase in the specific gravity during hunger peri- 
ods. J. Davy 4 obtained the same result by depriving his subject only 
of water, and Nasse, 5 starving dogs but allowing water, found that a 
decrease occurred in specific gravity after 3 to 4 days, but that by the 
eleventh day the blood had returned to or even exceeded its normal 
density. 

Liuboumdrow, 6 using the pyknometer, detected fluctuations in den- 
sity as marked, comparatively, as those noted in the number of red 
cells, but a complete agreement between specific gravity and erythro- 
cyte count was not found. 

Castellino, 7 studying starving rabbits, found an increase in density 
and at the same time a decrease in the serum content of their bloods. 

Popel 8 also reports an increase, though a slight one. He studied 
both rabbits and dogs, using Hammerschlag's method. In the former 
the increase did not exceed 1.6 per cent and it was still less in the dogs. 
After ligation of the ureters, there was the slight fall of 9.11 per cent 
from normal in rabbits, while the dogs showed a rise of 0.72 per cent, 
a rather unexpected result, if taking only the water content of the blood 
into consideration. 

1 Curran, The pathology of starvation, Med. Press and Circ, London, 1880, n. s., 29, pp. 
210 and 229. 

2 Jolly et Levin, Sur les modifications histologiques de la rate a la suite du jeune. Compt. rend. 
Soc. Biol. Paris, 1912, 72, p. 829. 

3 Thacrah, An inquiry into the nature and properties of the blood as existent in health and 
disease. London, 1819-1834. 

4 Davy, Physiolog. and Anat. Researches. London, 1839. 

B de Martigny u. Nasse, Ueber den Einfluss der Nahrung auf das Blut. Marburg u. Leipsic, 1850. 

•Liuboumdrow, Changes in the blood and organs in starvation. 71 Dissert., 1893, Russian. 
From the Path. Anat. Laboratory, Prof. W. Winogradow, St. Petersburg. 

7 Castellino, La Suscettibilita infettiva nella inanizione lenta. Riv. d'Igiene e Sanita Pub., 
Roma, 1893, 4, No. 3, p. 461. 

8 Popel, Sur les variations de la densite du sang dans le jeune absolu, simple, ou complique de 
la ligature des ureteres. From the Laboratory of General Pathology, Prof. S. Lukjanow, Arch, dea 
eci. Biol., 1895-96, 4, p. 354. 



THE BLOOD. 145 

London's 1 findings do not agree with those above. He also used 
Hammerschlag's method, but reports a slight diminution in the rabbits 
that starved for from 5 to 14 days, the average dropping from 1.048 
to 1 .043. The animals in both the above series were deprived of water. 

There was evidently a considerable fall in Martin's 2 blood, for while 
no preliminary estimation was made, on the sixth day of his fast the 
specific gravity was 1.026; on the eighth it rose to 1.031, and on the 
ninth and last day it had dropped to 1.021. One week after breaking 
fast it was 1.043, still very low if we consider the normal to be 1.059 to 
1.060. 

Lloyd Jones is quoted by Lyonnet 3 as finding, on the tenth day of 
one of Succi's fasts, a specific gravity of 1.061 that rose to 1.063 on the 
thirty-ninth day. In speaking of the influence of food and drink on 
the specific gravity, Lyonnet holds that there is usually, though not 
invariably, a diminution after the intake of water, the change being 
but very temporary. Abstinence from all liquid causes an increase, 
but not of so marked a degree as one would suppose. (In this he is 
quoting Lichtheim.) Food apparently has some effect, in that after 
meals there is a decrease to be found that lasts for an hour or so. 

Coagulability. — Very little mention is made in the literature of the 
influence of inanition on the coagulation time. Vierordt 4 was among 
the first to refer to this feature of the subject, having made the obser- 
vation, in 1878, that an acceleration of the process occurred as a result 
of starving. 

Arnold 6 and Collard de Martigny 6 ' 6 both noticed that the clot was 
larger than usual in relation to the amount of serum, and the latter in 
1850 found a decrease in fibrin content. 

Jones 5 also noted that the water and fibrin decreased more rapidly 
than the solid constituents. 

Kallmark 7 noticed that in rabbits, after the fifth or sixth day of 
starvation, the blood coagulated more rapidly, but no estimations of 
the time are given. 

Tria 8 reports quite recently that he could detect very little variation 
during short fasts in rabbits and dogs. 

1 London, Note sur la question du changement de la quantity generate et de l'alcalinite du sang 
dans le jeune absolu. Arch, des Sciences Biol., 1895-96, 4, p. 516. (Abstract by Miihlmann. 
See footnote 5, this page.) 

2 Gordon, A prolonged fast. Montreal Med. Journ., 1907, 36, p. 482. 

3 Lyonnet, De la density du sang, sa determination clinique, ses variations physiologiquea et 
pathologiques. Paris, 1892, p. 73. 

4 Vierordt, Arch. d. Heilk., 1878, 14, p. 193. 

'Miihlmann, Russisch Literatur tiber die Pathologie des Hungerns. Centralblatt f. allgem. 
Path., 1899, 10, p. 160. 

6 de Martigny u. Nasse, Ueber den Einfluss der Nahrung auf das Blut. Marburg u. Leipsic, 1850. 

7 Kallmark, Zur Kenntniss des Verhaltens der weissen Blutkorperchen bei Inanition. Folia 
Hsemat., 1911, 11, pt. 1, p. 411. 

8 Tria, Propriet6s chimico-physiques du sang durant l'inanition. Archiv. ital. de biol., Pise, 
1911, 55, p. 49. (Arch, di farmacol. sper., Roma, 1909, 8, p. 359.) 



146 A STUDY OF PROLONGED FASTING. 

Valentin 1 noticed a marked retardation of coagulation in hibernating 
animals. 

No report of the specific estimation of coagulation time during fast- 
ing in man could be found. Dr. Wile 2 reports that on both examina- 
tions of Gayer's blood, made on the eighteenth and thirtieth days, 
there was apparent decrease in the platelets, but that coagulation was 
accelerated. He says, of the last examination, that "the blood was 
thick, dark red, and did not flow easily." Aside from such general 
conclusions without data to demonstrate them, the only clue as to what 
might be expected in man are a few observations that have been made 
relative to meal times. 

Coleman 3 found the longest coagulation time an hour after the 
principal meal and the shortest before breakfast. 

Cohen, 4 using the method devised by himself, determined that the 
average time before meals was 7-g- minutes and after meals 9 minutes, 
while Mercier, quoted by Cohen in the above article, constantly found 
the coagulation more rapid after meals than before, and Addis 5 claims 
that food has no influence on the process. 

Cohen 4 quotes A. E. Wright as crediting fluids with a greater influ- 
ence on coagulability of the blood than food, but that hunger does 
retard the process, a view not upheld by the observations of Coleman 
and Cohen. Increased consumption of liquids lengthens the time and 
withholding them has the opposite effect. 

Immunity. — There have been a few studies made of the effect of 
starvation on immunity in general and the immune body-content of 
the blood specifically, but the data are scarcely sufficient to warrant 
definite conclusions. 

In 1890 Canalis and Morpurgo 6 studied the effect on the natural 
immunity pigeons exhibit toward anthrax. They were found con- 
stantly to lose this resistance if the fast were begun immediately after 
the injection of the organisms, or a day or so before. They regained 
it, however, on refeeding, if the inanition period had not been too long. 
This same natural immunity possessed by chickens 7 was not lost 
unless they were starved for more than 8 days. If starved before 
inoculation they proved more susceptible. These workers were unable 
to make rats susceptible to anthrax by starving. 

Valentin, Repel, f. Anat. u. Physiol., 1838, 3, p. 156. 

2 Gayer's fast : A private communication from Dr. Wile, of New York City. 

'Coleman, The coagulation of the blood and the effects of certain drugs upon it. Biochem. 
Journ., 1906-7, 2, p. 184. 

4 Cohen, Coagulation time of the blood as affected by various conditions. Arch. Int. Med., 
1911. 8, pp. 684 and 820. 

6 Addis, The coagulation time of the blood in man. Quart. Journ. Exp. Physiol., 1908, 1 , p. 305. 

6 Canalis and Morpurgo, Ueber den Einfiuss des Hungers auf die Empf anglichkeit fur Inf ections- 
krankheiten. Fortschr. der Medicin, 1890, 8, p. 693. 

7 Canalis and Morpurgo, ibid., 8, p. 729. 



THE BLOOD. 147 

P. Castellino 1 concluded, from his studies on rabbits in 1893, that 
there was a diminution in resistance to infection. 

A few years later, in 1899, Meltzer and Norris 2 could demonstrate no 
difference in the bactericidal action against the typhoid bacillus of the 
blood of starved, under- or overfed dogs. 

Roger and Josu6, 3 on the other hand, having observed an increase 
in the resistance to the colon bacillus in fasting rabbits, suggest that 
some possible benefit may be derived from fasts. This is the only bit 
of experimental evidence — with reference to the blood, at least — that 
speaks for the value or advisability of this procedure as a general thera- 
peutic measure. This increase in resistance they attribute to hyper- 
activity of the bone marrow, whereby there is a more rapid proliferation 
of the defensive cells. 

Charteris 4 noticed a wide daily variation in the opsonic index of the 
blood of his human subject during the latter's 14-day fast, but, as he 
obtained a similar result with his own blood, he was led to conclude 
that the changes during fasting were not significant, due, rather, to the 
use each day of a fresh emulsion of bacteria. Martin's 5 blood, however, 
showed a gradual lowering of the index, returning to normal 4 days after 
the fast was broken. In this case the Staphylococcus aureus was used. 

Bizzozero 6 studied the natural hemolytic power of the blood serum 
of 8 chickens that starved for from 8 to 17 days and could find practically 
no alteration. He concludes that the hemolysins are not concerned in 
the defense of the organism against bacterial invasion, because, as we 
have seen from the work of Canalis and Morpurgo, starving does lower 
the resistance to infection. 

Among the studies on other properties of the blood are to be men- 
tioned those of Tria, 7 on the viscosity and electro-conductivity in rab- 
bits and dogs. He found little alteration, some decrease, in both early. 
He concludes from his entire study that the body is able to compensate 
pretty well for the disturbances in nutrition, thus permitting of long 
fasts without serious consequences. The investigations of Deter- 
mann, 8 and of Maran6n and Saristan 9 dealt especially with the viscosity. 

Castellino, La Suscettibilita infettiva nella inanizione lenta. Riv. d'Igiene e Sanita Pub., 
Roma, 1893, 4, No. 3, p. 461. 

2 Meltzer and Norris, On the influence of fasting upon the bactericidal action of the blood. 
Journ. Exp. Med., 1899, 4, p. 131. 

3 Roger et Josue, Influence de l'inanition sur la resistance a l'infection colibacillaire. Compt. 
rend. Soc. Biol., 1900, 52, p. 696. 

4 Charteris, Record of changes observed in the blood count and in the opsonic power of a man 
undergoing a prolonged fast. Lancet, 1907, 2, p. 685. 

6 Gordon, A prolonged fast. Montreal Med. Journ., 1907, 36, p. 482. 

"Bizzozero, Pouvoir hemolytique naturel, pulet dans l'inanition aigu6. Arch. ital. de biol., 
Turin, 1904-5, 42, p. 212. 

7 Tria, Proprietes chimico-physiques du sang durant l'inanition. Archiv. ital. de biol., Pise, 
1911, 55, p. 49. (Archiv. di farmacol. sper., Roma, 1909, 8, p. 359.) 

8 Determann, Die Beziehung der Viskositat des Blutes zu den Korperfunktionen. Veroffentl. 
d. balneol. Gesellsch. in Berl., Berlin u. Wien, 1910, pp. 259 and 270. 

9 Marafi6n y Saristan, La viscosidad de la sangre humana en varios estados patal6gicos. Rev . 
Ibero Am. de cien. m£d., Madrid, 1911, 26, p. 244. 



148 A STUDY OF PROLONGED FASTING. 

A decrease in alkalescence was noticed by Tauszk 1 in Succi's blood, 
by Castellino 2 in rabbits, and Benedict reports the same change in his 
subject. 3 A very moderate decrease was also observed by London 4 in 
his eight rabbits. Castellino 2 found also a decrease in NaCl content 
and in the bulk of serum. 

For additional data as to the effects of inanition on the physico-chem- 
ical properties, reference can be made to the work of Githens, 6 Schce- 
neich, 6 Fria, 7 Lattes, 8 Robertson, 9 Bierry and Fandard, 10 Daddi, 11 Moroz- 
off, 12 and Weber 13 (who includes an exhaustive correlation of references 
to the literature of the entire subject of inanition). Manca 14 and Mac- 
alum 15 confined their investigations to the cold-blooded animals. 

OBSERVATIONS ON L.*S BLOOD. 

There is little danger of one's opinions being biased by the diverse 
results above correlated. We can therefore take up the consideration 
of our subject's blood either with an open mind free from preconceived 
ideas, or with confused expectations, ranging from absolutely negative 
findings to very grave disturbances, with the confidence that we have 
precedent for almost any picture that may present itself. The coagu- 
lation time and specific gravity were investigated, but the examina- 
tions were concerned principally with the red and white cell and haemo- 
globin content, the technique for which follows. That employed in the 

tauszk, Jahrsb. tiber d. Fortschr. der Thier-Chemie, 1894, 24, p. 147, abstracted from Orvosi 
hetilap, Budapest, 1894, p. 512 : also Hamatologische Untersuchungen am hungernden Menschen. 
Wien. klin. Rundschau, 1896, 10, p. 306. 

2 Castellino, La suscettibilita infettiva nella inanizione lenta. Riv. d'Igiene e Sanita Pub., 
Roma, 1893, 4, No. 3, p. 461. 

3 Benedict, Carnegie Inst. Wash. Pub. 77, 1907, p. 322. 

4 London, Note sur la question du changement de la quantity g£nerale et de Palcalanite du sang 
dans le jeune absolu. Arch, des Sciences Biol., 1895-96,4, p. 516. (Abstract by Miihlmann. See 
footnote 5, p. 145.) 

KJithens, Influence of hunger and haemorrhage on the composition of the blood plasma. Proc. 
Phila. Count. Med. Soc, Philadelphia, 1904-5, 25, p. 279. 

6 Schosneich, Beschaffenheit des Blutes unter verschiedenen Bedingungen. Ztschr. f. exp. Path, 
u. Therap., 1905, 2, p. 419. 

7 Fria, Alcune ricerche comparative sul sangue di animali nutriti naturalmente ed innaturalmente. 
Folia clin., chim. et micros., Salsomaggiore, 1910-11, 3, p. 44. 

8 Lattes, Ueber den Fettgehalt des Blutes des Hundes unter normalen u. unter verschiedenen 
experimentellen Verhaltnissen (Verdauung, Hungern, etc.). Arch. f. exp. Path. u. Pharmacol., 
Leipzig, 1911, 66, p. 132. 

9 Robertson, Studies in the blood relationship of animals, etc. 1. A comparison of the sera of 
the horse, rabbit, rat, and ox with respect to their content of various proteins in the normal 
and in the fasting condition. Journ. Biol. Chem., Baltimore, 1912, 13, p. 325. 

10 Bierry et Fandard, Variations de la glycemie pendant l'inanition. Compt. rend. Acad. d. sc, 
Paris, 1913, 156, p. 2010. 

u Daddi, Surles modifications du poids de l'extrait 6th6r6 du sang durant le jeftne de longue 
dur6e. Arch. ital. de Biol., Turin, 1898, 30, p. 439; also Sulle modificazioni del peso dell estratto 
estereo del sangue durante il digiuno di lunga durata. (Sperimental.) Arch, di biol., Firenze, 
1898, 52, p. 43. 

^Morozoff , On the effect of fasting for a short time on the morphologic composition of the blood. 
(Russian) Vrach. St. Petersburg, 1897, 18, p. 1081. 

13 Weber, Ueber Hungerstoffwechsel. Ergebnisse der Physiologie (Biochemie), 1902, 1 Abt., 
p. 702., 

14 Manca, Le cours de l'inanition chez les animaux a sang froid. Arch. ital. de biol., Turin, 
1895, 23, p. 243, and 1896, 25, p. 299; also Chemical researches on animals (cold-blooded) during 
inanition (Italian). Arch. ital. de biol., Turin, 1903, 39, p. 193. 

15 Macalum, The inorganic composition of blood plasma in the frog after a long period of inani- 
tion. Rep. Brit. Assoc. Adv. Sc, London, 1911, 80, p. 766. 



THE BLOOD. 149 

coagulation and specific-gravity estimations is given under these head- 
ings. The attempt was made to determine the opsonic index, but the 
subject was so far from an incubator and centrifuge that it was found 
impossible to obtain accurate results. There was a possibility, also, of 
this additional manipulation having a disturbing influence on the physi- 
ological investigations, so this feature was abandoned, though still recog- 
nized as one of the most important lessons to be learned from the blood. 

Technique. — For the three days just preceding the fast, Levanzin's 
blood was examined to determine a normal picture with which to 
compare the results later obtained. During the fast, with the excep- 
tion of the first day and three days scattered through the period, 
daily examinations were made and also on the first and third days of 
refeeding. The time of the day did not vary more than half an hour 
throughout, all the specimens being obtained between 10 and 10 h 30 m 
a. m., so that there was a constant relation to the general routine of 
the subject's daily activities — that is, immediately after he had finished 
with the respiration experiment, had been weighed, had washed his 
face and hands, and climbed the short flight of stairs to his balcony. 
It was his habit to take about half a glass of water before submitting 
to the lancet prick. (It may be well to mention here that, for the first 
10 days of the experiment, the subject received the constant quan- 
tity of 750 c.c. of distilled water per day, and thereafter, 900 c.c. 
per day.) The specimens of blood were obtained from alternate 
fingers of the left hand and occasionally from the ear. Deep pricks 
were made, so as to obtain sufficient blood without squeezing. For 
counting the red and white cells the Thoma-Zeiss apparatus was 
used, diluting with fresh salt solution for the former and 1 per cent 
acetic-acid solution tinged with gentian- violet for the leucocytes. The 
usual precautions were taken to insure uniform suspension of corpuscles 
and the even filling of the counting chamber. In the case of the ery- 
throcytes, 80 small squares were counted and 4 ciphers added to the 
total. The average of two or more such figures was taken as the final 
result. In estimating the leucocytes, the whole cross-ruled field was 
counted and the result multiplied by 200. The average of three or 
more such figures was taken as the final estimation. The use of gentian- 
violet in the acetic-acid solution makes the counting of the leucocytes 
much easier and the likelihood of mistaking foreign particles for cells 
practically impossible. The smears for the differential counts were 
stained by the Wright method; 200 or more cells were examined 
by the use of the tV objective and No. 1 eye-piece, and classified as fol- 
lows : Polymorphonuclear neutrophile, eosinophile, and basophile ; small 
and large lymphocyte ; monocyte ; transitional cell. As the classification 
of the last three forms is so mooted a question, it will be necessary to 
go into some detail as to just what cells were placed under these heads. 

Under large lymphocyte was classified the mononuclear cell, con- 
siderably larger than the red corpuscle, with a round or typically 



150 A STUDY OF PROLONGED FASTING. 

indented nucleus (like that of the small lymphocyte though not staining 
as deeply), small amount of cytoplasm in proportion to nucleus, the 
former not being markedly basophilic and usually containing a few 
faintly-staining granules. The mononuclear was considered the cell 
whose nucleus was more indented, the proportion of cytoplasm was 
greater and granules more evident, the latter basophilic and the whole 
cell staining more deeply than the large lymphocyte. The occasional 
large mononuclear cell of the endothelial type was also counted in with 
these cells, though not considered as being associated with them generi- 
cally. The transitional cell showed a pale-staining, usually kidney or 
horse-shoe shaped nucleus, surrounded by cytoplasm, pale and free from 
granules. This classification follows generally that of Pappenheim. 1 

In estimating the haemoglobin percentage the Tallqvist 2 scale was 
used throughout the series. This method, it is true, is open to some 
criticism in that it is scarcely possible to detect differences of less than 
3 per cent. My experiences, however, have developed a confidence in 
it that has been justified by comparative readings with other methods. 
In the present case, from the nineteenth to the twenty-fourth day, 
estimations were made with the Sahli apparatus and the results were 
practically the same as those obtained for the same days with the 
Tallqvist scale. Care must be taken to follow exactly the same tech- 
nique for each reading, particularly in the matter of light; the results 
will then be relatively correct, even if the method is comparatively 
weaker than some of the others. 

The results of these examinations are contained in table 16. They are 
correlated and presented more graphically by curves given in figures 
20 and 21 . No. I shows the relation of haemoglobin to red cells, and No. 
Ill the relation of the total white-cell count to the differential. To avoid 
confusion of curves the transitional, eosinophile, and basophile are 
plotted separately in No. IV and the scale enlarged. In No. II the 
composite curve of the polynuclears, that is, neutrophiles, basophiles, 
and eosinophiles and one of the mononuclear cells, large and small 
lymphocytes, transitionals, and monocytes, are given for comparison. 

Erythrocytes. — The subject's normal count apparently was high; the 
three preliminary estimates range well above 6,000,000. It maintained 
this high figure throughout the test, going below it on only two occasions, 
the tenth day of the fast and the third day following. In the early part 
of the fast there is daily variation, ranging under 1,000,000. This 
becomes less evident toward the end. The general impression given 
by the curve is that of a very moderate decrease. There were no alter- 
ations in the characteristics of the individual cells as to size, shape, and 
staining properties and no nucleated red cells were found at any time. 

1 Pappenheim u. Ferrata, Ueber die verschiedenen lymphoiden Zellformen des normalen und 
pathologischen Blutes. Folia Haemat., 1910, 10, p. 78. 

2 Tallqvist, Ueber die Anwendung des Filtrirpapiers in Dienst der praktischen Hsematologie, 
Berl. klin. Wochenschr., 1904, 41, p. 926. 



THE BLOOD. 



151 



Haemoglobin. — The haemoglobin average ranges rather consistently- 
above 85 per cent, the most marked variations being consistent with 
those of the erythrocytes. The low period is between the tenth and 
sixteenth days, from then on showing a very moderate rise. 

Table 16. — Levanzin's cell counts and haemoglobin percentage. 



Days. 



Haemo- 
globin. 



Total 
erythro- 
cytes. 



Total 
leuco- 
cytes. 



1 

5 



p^ 



gjj 

>> 

a 

CO 









PQ 



\_ 



Days before fasting: 

3d 

2d 

1st 

Days of fasting: 

2d 

3d 

4th 

6th 

6th 

__7th 

9th 

10th 

11th 

12th 

13th 

14th 

16th 

17th 

18th 

19th 

20th 

21st 

22d 

23d 

24th 

, ^25th 

^ 27th 

28th 

29th 

30th 

31st 

Days of diet: 

1st 

2d 

3d 



p. ct. 
85 
85 
95 

90 
92 
92 
91 

90 

88 
90 
88 
85 
85 
85 
87 
85 
88 



87 
88 
90 
88 
87 
90 

90 
92 
93 

92 

92 



8,984,000 
6,632,000 
7,000,000 

6,100,000 
7,200,000 
6,120,000 
7,170,000 
6,250,000 
6,450,000 
7,000,000 
5,950,000 
6,750,000 
6,480,000 
6,580,000 
6,280,000 
6,010,000 
6,600,000 
6,700,000 
6,250,000 
6,250,000 
6,450,000 
6,130,000 
6,630,000 
6,000,000 
6,250,000 
6,240,000 
6,350,000 
6,190,000 
6,050,000 
6,170,000 



5,900 
6,400 
6,000 

8,400 
12,400 
8,400 
9,400 
8,600 
7,000 
8,600 
7,000 
8,400 
7,000 
6,800 
7,900 
9,000 
6,200 
6,600 
6,200 
6,400 
6,400 
6,100 
6,600 
6,900 
6,500 
5,800 
7,800 
6,000 
6,000 
8,000 



p. ct. 
67.5 
66.0 
59.5 

73.0 
79.0 
66.0 
68.0 
67.5 
60.0 
66.0 
68.0 
68.0 
74.0 
64.0 
64.0 
72.5 
64.5 
65.5 
62.5 
61.0 
62.5 
61.0 
58.5 
65.0 
65.5 
63.0 
61.5 
63.5 
64.5 
60.0 



6,280,000 7,200 70.0 

Too ill for examination. 

5,960,000 I 6,600 | 60.5 



p. ct. 
23.5 
22.0 
32.0 

20.0 
16.5 
24.0 
26.5 
19.0 
24.0 
26.0 
22.0 
23.0 
15.0 
24.5 
26.5 
16.5 
29.0 
24.0 
22.5 
28.5 
20.5 
26.5 
28.0 
25.0 
27.0 
26.0 
30.0 
27.5 
26.5 
32.0 

21.5 

32.5 



p. ct. 
2.5 
1.5 
3.0 

3.0 
.6 
4.0 
2.0 
4.5 
4.0 
3.6 
3.0 
2.0 
2.0 
7.6 
4.0 
3.5 
3.0 
6.5 
5.5 
3.0 
5.0 
3.6 
4.5 
2.5 
1.5 
3.0 
1.0 
1.5 
2.0 
2.5 

2.5 

3.0 



p. ct. 
4.5 
7.5 
4.0 

3.0 
3.0 
4.0 
3.0 
3.0 
3.5 
1.0 
3.0 
5.0 
4.5 
1.0 
2.6 
1.5 
.5 
2.0 
3.6 
3.5 
6.5 
6.5 
4.5 
4.5 
6.0 
6.5 
5.5 
5.0 
5.0 
5.0 

4.5 

3.0 



p. ct. 

1.5 

2.0 





1.5 




p. ct. 

0.5 

.5 

1.5 

. 5 

.5 

.6 



.5 

2.0 

.5 

1.0 



1.5 

.6 

.5 

.5 



1.0 

.5 

1.0 

1.5 



.6 

.6 









.5 





.5 



p. ct. 

.5 







.6 



.6 



.6 


.6 



.5 

.6 


1.0 





.5 









Leucocytes. — It is in the total leucocyte count that we have the most 
striking change of the whole series of examinations. There was a rapid 
rise at the onset of the fast, reaching 12,400 on the third day. On the 
fourth day, however, it immediately fell to 8,400, after which there was 
a consistent daily variation of about 1,000 until the sixteenth day, 
when it reached approximately the preliminary count, after which there 
will be noticed a more marked daily variation. It was only possible 
to continue the examinations for 3 days following the fasting period, 
so that the count had not settled down to normal when the subject 
went out from under observation. 



152 



A STUDY OF PROLONGED FASTING. 



APRIL MAY 

II 12 1314-15 16 17 1819 2021 22 23242S 2627 282930 I 2 3456 7 S 9 101 1 12 I3I4I5I6I7 



DAY 3 


2 


1 


I \Z 


3 


4 


5 


6 


7 


S 


9 


10 


II 


12 


13 


4 


5 


6 


7 


8 


19,2021 


222324 


25 


26 


2282919 


31 


112 3 














































































































































































































7,000,000 






l 


/ 






















































i 










/ 


re 


/ 


\ 


/ 








/ 


, 












































ERYTHROCYTES 




/ 


v 


/ 


\ 


/ 


\ 












\ 
















































-fc 


/ 


\ 




\ 


f 


' 






/ 


\ 


^ 


S 






/ 




\ 








/ 


























IE 


/ 


\ 


/ 




/ 






\ 


/ 






s 


N 




/ 




\ 




7\ 


\ 




\ 






■* 




s 






X 


6<000,000 






i \ 
















1 


1 










N 


' 










^ 




\ 


/ 










--> 


^ 


'\ 




































































06* 






































































/ 


v: 


































































/ 


x 






■~s 


















































:^! 








p 


/ 






S 


\ 




S 


s 








































r 






HEMOGLOBIN 




1 


i 










^ 


r" 




^ 


1 












































65% 






i 


















\ 






/ 


«v 


i 


/ 




















































































































































































































































































































































































BE 


FC 


Rl\ 


IN 


A 


in 


10 


N 


PC 


R1C 


D 


+01ET 



DAY 3 


? 


1 


HISSSSZ 


3Si5u.I2i3!4i5 


gi7igi9 2S 


21222324 


jiESSI 


3303! M2 3 








1 -t 
















































9,000 
















T i 








1 1 










JL i 








1 


















1 1 


















1 












8,000 






1 1 


















1 


















1 1 
















































7,000 






1 


















1 


















1 


















1 




J 














1 A 




4 I 








6,000 






tu tX 




tT I 














T7 t^ 


















M- 


J^L / 
















4t L. 


4t4A t 
















J- t 


txt-Z t 










5,000 






x i 


AL X—T 








1 








i 4 


1 XI 








- ZK 








x H 


t At 






j[ 


- 44A 








4^ H 


r -*f- 




_ _j 


V 4 


, 4^V 


POLY NUCLEAR 


M 




ft 






,_ 7 


V 4- 


\ t ! A 


4,000 


V 


tt. it 




Z!SI 


21 £ 


A-t- 


t ! 








tt 






~xu 


VL- 


- 






1 


i 








\ 








































3.000 
















- Jl ' 
















4\ 


- 4W-- 










7 rr is 






2 


h 44^- 










V"7r / 






I. 


k ft^ 










J£-V4 






^TX- 


V V 


2.000 








W 








- r 










-¥- 










MONO NUCLEAR 








T 














































1,000 




















3£F0 












!Ef 







Fig. 20. — Chart I. Relation of haemoglobin to erythrocytes. 
Chart II. Composite curve of the polynuclears 
compared with one of mononuclears. 



THE BLOOD. 



153 



APRIL MAY 

II 12 13 14- 15 16 17 18 I9Z02I 222334252627282930 I 2 3 4 5 6 7 8 9 10 II l2 1314 15 16 17 



DAY 3i 


J I 


ii2 34: 


5 6 7 8910, 


1 1 12 13 


141516171819$ 


2i222J 


24 


25 


26272 


8 29303 


1 112 3 






























1 2,000 
































1 




























1 




















1 








1 




















1 








1 
























1 1,000 




1 T" 




























1 -i- 


































' - 






















1 1 
























lO.ooo. 




1 




























1 




























1 




























1 




























1 


jl i 






















9,000 . 




1 II 


tX 


























1 4U 


.% 






/ 




















1 111 


j 






/ 




















1 tV- 


J 






i 






















7 


j 




/ 
















8,000 




Ml". 


- 1- tl 




n 






















mX - 


- 4 LI 


/ 


















1 1 






nrt : 


- 44 4 


/ \ 


1 
















\l 






it - 


- - J 4 


/ \ 


7 














1 


v 






it-- 


- t J 


/ \ 


r 
















V 


7,000 . 




±t 


r 


' 


t 
















k 






j~ ... 




s 










/ 


, 




1 


i s 






jt .. 














/ 


\ 












i -- 


JE 






i 


A , 


/ 




1 










A 


tr j. 


t - 






/ 


/ \/ 


\ 












TOTAL 6,000 1 


\ 


_T J 


i 






/ 




A^ 






V. 






LEU KOCYT ES — * - 




i / 


■j - 






; 










\ 










ZlI 








/ i 




















it . 


- ^c 


— IL- 




' 




















it . 


- - tx 


\ 




















5,000 . 




( 


- x t\ 


/ I 


J 






















i . 


- 44 J 


/ \ 


44 




\ 












A 






i- - 


- -4 -1 


' \ 


I 














■JU - 


i \ 






jr . 


- t 


\ 


£- 














A 


'\ 






IE . 


. t _ 








a 




/ 


\ 




\ . 


i V 


4.000 1 


\ 


tr . 










t V 




/ 




V 


\ / 


i > 


POLY NEUTRO-"*- 


\ 


ft . 










^ 


A. 


f 




V J 


i ' 






1 


□C - 












^7* 






\f 






























































3.000 - 






















































































































iu. _, 


















ji_ 


ct 


2.000 . 




i ~Z. 


-C Zw 


















/\ i 


'\' / 






"4 / 


-A v \ 


A 


7^ 












1 


\ / 


v / 




/ 




-J Z. 4 


f i 


L. 


V, 


/N > / 


^ / 


■^ 




*>J 




-42 


SMALL LYMPH. 


/ 






\ / 




\' 


V 


\/ 






^v.' 




V 










\/ 




















1,000. 








V 












































































LARGE MONO-n. , 


K 






-, — -* 


t 


















LARGE. LYMPH-J. 


\ 


mkrfl' 






- 


" — h 




7<> 


S" 




i 






o: 


~K 


1 " x ' 




~-f\ 






*,*-*" 








_a»c 


*—- ' 


' _ r~* 


DAY 3 


> 1 


1 12 34 J 


> 6 7 8 9 10 


II 12 13 


14 1 


3161 


7*1819 20 


2J2223 


?A 


?5 


2627? 


129303 


1112 3 








- Jti- 




























-Z-t 






j 
















4-00. 






--1 4 






/ 






















-4 4 






/ 












\/ 










-4 4 






/ 












V 








"t n 






/ 






















-4 L 






















300. 






-4 t 






1 






















-t 4 






/ 






















-t 4 








1 li 




















r i 




; 




1 




























\ 














200. 






: l 




























t 


A 


31 




t 


/ 


\ 
















, i 


1 \ 


z: 




-i-4 


/ 


















1 A 


/ \ 


t- 










1 












1 J 


r !J "i 


/ y 


r~ 




\ 






\ 




(\ 




100 , 




i 7 




1 






\ 






\ 


/ 


\ 




TRANSITIONAL I 


1 


, i ' 




i 






\ 


, 




\ 




y 


A 




y 


X / 




i r 






H 


cX. 






/ 


\ 


/\ 


EOSIN^. 






..J - 'r 


TT't 1 




.. 




/ 










VI \ 


BASO-s ; 


i\ \ 




"A V 






/I 


t i-v~ 


s • / 


\- 








1 V 


o\ 






^=-^ 


nIV 


c; 


1 


O 


VJP 


\ 






£N 


•,..i-- \ 



Fig. 21. — Charts III and IV. Relation of total to differential leucocyte counts. 



154 A STUDY OF PROLONGED FASTING. 

Polymorphonuclear neutrophiles. — These ran throughout rather con- 
sistently with the total count and the marked variations in this latter 
were quite apparently due to the change in polynuclear content. 

Small lymphocytes. — There are fluctuations in the count of this cell, 
except during the period between the fourteenth and twenty-seventh 
days, when they were comparatively constant, as were also the total and 
polymorpho-neutrophile counts. It will be noted that the majority of 
the rises and falls are the opposite of those seen in total and polymor- 
pho-neutrophile curves. This is particularly the case on the second, 
fourth, thirteenth, fourteenth, sixteenth, seventeenth, twentieth, and 
twenty-first days of fast and the first day of refeeding. It seems safe 
to conclude, therefore, that these fluctuations are only relative, due 
really to the fluctuations in polymorphs, and that their number was 
practically constant throughout the fast. 

The other forms of leucocytes present no distinctive features except the 
transitional, which was subject to several rises, namely, on the seventh, 
sixteenth, nineteenth, and twenty-first days, when they were above 5 
per cent. Only an occasional eosinophile was found during the last 

10 days of the fast and they had not returned to their usual number 
when examinations were discontinued. By examination of chart No. 

11 (figure 20), there appears to be a very slight increase in the combined 
mononuclear cells throughout the fasting period, the average being 
raised by the fluctuations in the transitional form, for the other types, 
i. e., monocyte, large and small lymphocyte are practically constant 
throughout, except the variations already noted in the latter. 

Coagulation time. — Toward the end of the second week of inanition 
it was noticed that the blood coagulated more rapidly than it had during 
the earlier days. This became more noticeable each day, so that if the 
mixing pipettes were not filled very rapidly the drop would coagulate 
or the blood would clot in the tubes. It is certain that this was not due 
to any physical alteration of the patient's environment. The tempera- 
ture of the balcony where the subject stayed and where the estimations 
were made was practically constant. This is a very important factor, 
for most experimental evidence goes to show that variations in temper- 
ature have decided influences on the coagulation time. Addis, Fox, 
and Wright, quoted by Cohen, 1 and the latter himself, all showed that 
rise in temperature accelerates and cold retards the process. Hartmann 2 
also notes that the higher the temperature the shorter the coagulation 
time, and Rudolf, 3 determining the effect more specifically, states that 
in general each degree of rise and fall between 15° and 20° C. decreased 
and increased, respectively, the time one minute. 

On the seventeenth day the estimations of the coagulation time were 
begun. Until the twenty-fifth day the McGowan 4 method was used. 

1 Cohen, Coagulation time of the blood as affected by various conditions. Arch. Int. Med., 
1911, 8, pp. 684 and 820. 

2 Hartmann, Zur Frage der Blutgerinnungszeit. Munch, med. Wochenschr., 1909, 56, p. 796. 

"Rudolf, Tr. Assoc. Am. Phys., Philadelphia, 1910, 25, p. 504. 

4 McGowan, A clinical method for estimating the coagulation time of the blood. Brit. Med. 
Journ., 1907, 2, p. 1580. 



THE BLOOD. 



155 



Capillary tubes of uniform caliber were filled with blood escaping after 
the specimens for the other examinations had been obtained, practically 
always the same relative drop, the third. Small sections were broken 
off at intervals of from 10 to 30 seconds. When a fine filament of 
fibrin was observed between the carefully separated ends of the tube 
and fragment, the time elapsed since the appearance of the drop used 
was taken as the reading. For the remainder of the examinations, the 
Boggs 1 apparatus was used. This consists of a truncated cone of 
glass that sets into a chamber into the side of which a small metal tube 
is inserted, connected with a rubber bulb, in such a way that a stream 
of air can be directed against the blood that is placed on the polished 
undersurface of the cone. As soon as coagulation has occurred in the 
drop, the mass of corpuscles that moved readily in one direction under 
agitation of the air-current merely vacillate. This method is more 
accurate than the first procedure, but is open, as are all the other 
devices for measuring coagulation time, to sources of error that, if not 
avoided, will cause wide variations in results. This refers particularly 
to the matter of temperature, the particular drop used, and the pres- 
ence of foreign particles in the blood or on the receiving surface of the 
cone, such as hair or lint. The first drop appearing after the prick will 
clot much slower than the subsequent ones, when the platelets will have 
accumulated about the edge of the wound. It is necessary, therefore, to 
use the same relative drop on each examination, preferably the second, 
though, as has been stated, in this case the third was used. 

Table 17. — Coagulation limes of Levanzin and a control. 



Day. 


Levanzin. 


J. F. 


Remarks. 


17th 


1' 5" 


0' 3" 


McGowan method. 


18th 


1 20 


2 50 


Do. 


19th 


1 20 


2 00 


Do. 


20th 


1 5 




Do. 


21st 


55 

1 20 
1 5 
1 5 


1 55 

1 35 

4 20 


Do. 
Do. 
Do. 

Boggs used on control. 


22d 


23d 


24th 


25th 




4 50 


From this day on, Boggs 


27th 


3 20 


4 + 


used on both. 


28th 


2 50 






29th 


2 20 






30th 


1 50 


3 30 




31st 


2 30 
2 45 


6 20 




First diet 



Note: '= minutes; " = seconds. 

The figures as obtained by these two methods, together with those 
found in a normal individual examined on the same days with the same 
methods and under approximately the same conditions, are collected in 
table 17. The last record on the control was made in a temperature at 

'Boggs, Johns Hopkins Hosp. Bull., Baltimore, 1904, 15, p. 174. 



156 A STUDY OF PROLONGED FASTING. 

most 2° C. cooler than that of the subject's environment, but a deduc- 
tion of 2 minutes, as correction for this, still leaves the time distinctly 
longer than that obtained on the same day in Levanzin's blood. A 
comparison of the two series of results will demonstrate a distinct 
increase in the coagulability of the starving man's blood, more notice- 
able toward the end of his fast. 

Specific gravity. — The specific gravity was determined only twice, 
while the subject was eating his first meal after fasting and on the 
third day of refeeding. The first time it was 1.0612, the second it was 
1.0618. The estimations were made by the Hammerschlag 1 method, 
the specific gravity of the mixture of chloroform and benzol being 
determined by the pyknometer. As no figures were obtained, either 
before or during the fast, these two examinations are of little value, 
except that from them it may be assumed there is a very slight in- 
crease in density, taking 1.059 to 1.060 as the average normal specific 
gravity. 

DISCUSSION AND CONCLUSIONS. 

The results of the above studies are conspicuous rather from the 
absence than the presence of striking alterations in the blood picture. 
Really the only prominent features are the early rise in polymorpho- 
nuclear neutrophiles and the decrease in coagulation time. The leuco- 
cytes, i. e., the neutrophiles, at least, are the most sensitive of the blood- 
cells to changes in body conditions, and we know that apparently slight 
disturbances will call forth a recognizable increase in these cells — a cold 
bath, for example. They seem always to be on the alert, ready at the 
least evidence of disturbance to rush forth in defense of the organism. 
It is scarcely to be wondered at, therefore, that in response to such an 
unusual condition as starvation there should be an outpouring of the 
reserve supply, at least for a day or so, or until the organism has had an 
opportunity to adapt itself to the altered conditions. The variations 
in water-content of the blood can not be considered a factor in this rise, 
involving as this does only the one form of cell. The only explanation 
that suggests itself, therefore, and frankly not a particularly scientific 
one, is this alertness of the polymorpho-neutrophile and its ever-readiness 
to be on the defense for the organism. The products of the somewhat 
perverted metabolism may excite them into this early activity and 
later fail to do so, but there is no evidence to prove this supposition. 
It is not easy to understand why they should respond to toxic products in 
the early days and not during the later as well, unless we assume they 
acquire a tolerance for them, which seems improbable when we compare 
the reaction to infections, in which their fight is evident throughout the 
disease if the organism is to conquer. It is further possible that an insig- 
nificant, obscure source of bacterial infection happened to develop at 
this particular time; if so, there was no other evidence. 

hammerschlag, Wien. klin. Wochenschr., 1890, 3, p. 1018. 



THE BLOOD. 157 

As to the effect of the starving on the total quantity of blood, it does 
seem evident that there are fluctuations, at least in the first two weeks. 
By comparing the curves of the white- and red-cell counts it will be noted 
that the variations are synchronous; that on the third, fifth, ninth, and 
eleventh days particularly the noticeable increases in the one are accom- 
panied by equally frank rises in the other. It would seem that this 
could only be due to variations in water-content. The specific gravity 
would have gone far toward proving this point, but unfortunately this 
was not determined during this period. Taking these fluctuations as 
indication of variations in water-content, it appears that during the mid- 
dle period of the fast, at least, the equilibrium of intake (including that 
drawn from tissues) and output was pretty well established. The last 
counts made, namely, on the third day after the fast was broken, are 
considerably lower than those of the last day of fast. This no doubt is 
a relative decrease, due to increase in water-content. While the diet 
was of course limited, there was an increase in the intake of fluids. 

The haemoglobin appears to be particularly resistant, the percentage 
on the last day being within 2 per cent of the highest estimation, 
found on the day before fast began, though there was a moderate 
decrease during the second 10 days. 

It is difficult to account for the only other marked change in the 
blood — the acceleration of coagulation. Loss of water could be respon- 
sible, but there is no evidence that this occurred. The very slight 
increase in density at the last, determined by the specific gravity, would 
certainly not demonstrate a sufficient concentration. There were no 
estimations of the platelet content made, but it is possible that the 
explanation lies with these. An increase in them could be responsible. 

The final conclusions as to the effects of uncomplicated starvation on 
the blood to be drawn from the results of examinations on Levanzin are: 

1. There is a slight actual loss in haemoglobin, more marked during 
the second 10 days. 

2. There are moderate fluctuations in water-content, particularly 
during the first half of the period, and an increase after breaking fast, 
evident till after the third day, at least. 

3. There is a decided rise in polymorpho-neutrophiles in the early 
days. 

4. There is an increase in coagulability, especially after the first two 
weeks. 

5. In an otherwise normal individual, whose mental and physical 
activities are restricted, the blood as a whole is able to withstand the 
effects of complete abstinence from food for a period of at least 31 days, 
without displaying any essentially pathological change. 



MECHANICS OF RESPIRATION. 

A physiological study of the human body during a prolonged fast 
would be incomplete without a careful investigation of the influence of 
inanition upon the mechanics of respiration. Fortunately, it was pos- 
sible to obtain such data in all of the experiments with the respiration 
apparatus, as the spirometer gave a graphic record of the respiration, 1 
from which accurate data regarding the respiration-rate, the ventilation 
of the lungs per minute, and the volume of air per inspiration could be 
obtained. Such data are available for the morning respiration experi- 
ment for every day of the fast, for the experiments made in the evening 
before the subject entered the bed calorimeter, for experiments made 
on several occasions when the subject was sitting quietly or sitting 
writing, and also for experiments in which he breathed an oxygen-rich 
atmosphere. 

TYPICAL GRAPHIC RECORDS OF RESPIRATION. 

The graphic records obtained by the spirometer method have a 
special interest in connection with the fasting experiment in that they 
indicate the character and rate of the respiration as the fast progressed. 
Out of 200 or more records obtained with this subject, four typical 
curves have been selected for reproduction in figure 22, i. e., one each 
for April 17 and April 30, and two for May 14, 1912. From these 
curves it will be seen that at each inspiration the pointer on the spiro- 
meter rises and at each expiration falls. The experiments were so 
conducted that the communication between the subject and the spiro- 
meter was made at exactly the end of a normal expiration; consequently 
the first deviation from the straight line is that due to an inspiration. 
Similarly, at the end of the experiment the communication with the 
spirometer was cut off at the exact end of the normal expiration. 
From this record the respiration-rate can easily be counted. 

Immediately below the record of the respiration is the line showing 
the time in minutes; the lowest line indicates the number of revolutions 
of the recording device — the so-called work-adder wheel — from which 
the total volume of ventilation is calculated. Since a record of the 
muscular activity is essential for all intelligent comparison of the results 
of respiration experiments, a method was followed similar to that used 
for the bed calorimeter, the bed upon which the subject lay being 
provided with a pneumograph, tambour, and pointer, by which a record 
of the degree of muscular repose was obtained. This record is shown 
in the line directly above the respiration curve. Frequent testing has 
shown that this form of bed 2 is extremely sensitive. 

1 See description and schematic outline of spirometer on p. 317 and figure 40. 
2 See arrangement of bed inside respiration calorimeter, figure 37, page 312. 

158 



MECHANICS OF RESPIRATION. 



159 



In the curve for April 17, 1912, which was obtained near the beginning 
of the fast, it will be noted that the subject took a deep breath every 
2 or 3 minutes, but in general the vertical height of the various lines 
indicates a fair regularity in the volume of air inspired. The record 
of the degree of muscular repose shows that during the whole period of 
15 minutes the subject did not make a movement which could be 
recorded. As the recording device is so sensitive, it can be confidently 
asserted that the subject was in absolute muscular repose throughout 



I. PERIOD 2. APR. 17, 1912 




lllttltflkiiiiiiiiiw 



TIME IN MINS. 



VtNTILATION 



J 



llfflWIIHWinhiii.lli 



E. PERIOD I. APR. 3ft 1912, 

HI 



mmmm 



jm 



m. PERIOD 3. MAY 14. 1912 



***timmmtmm**Hm 



JZ. PERIOD 4. MAY 14. 1912 




Fig. 22. — Specimen respiration curves for subject L. when lying on couch in experiments 
with the respiration apparatus. 



160 A STUDY OF PROLONGED FASTING. 

the period so far as external muscular activity is concerned, although 
it is obvious that no idea of the muscular tonus can be obtained by 
this method. 

The second curve, that for April 30, 1912, was obtained about the 
middle of the fasting period and is typical of many obtained about this 
time. In this respiratory record but two abnormally deep breaths 
are noted. 

The third curve was obtained on May 14, 1912, at the end of the 
thirtieth day of the fast. In this record a greater frequency of respi- 
ration may be noted, with less amplitude, this being clearly shown even 
without measurement. The great sensitivity of the device for record- 
ing the degree of muscular repose is shown in the original kymograph 
curve by a wave-like line above the respiration record indicating the 
slight disturbance in the center of gravity of the body due to the res- 
piratory movements . While thi s may be very plainly seen in the original 
curve, it is lost in the reproduction. 

Immediately after the third curve was obtained, the apparatus was 
rilled with pure oxygen, so that the subject breathed an atmosphere 
containing 95 per cent oxygen. One deep respiration is shown in the 
curve obtained (curve IV). The rate is apparently a little slower than 
in the preceding curve and the volume somewhat larger. The line 
above the respiration record again shows that the subject was abso- 
lutely quiet throughout the whole period, as was usual with this man. 

METHOD OF CALCULATING THE TOTAL VENTILATION OF THE LUNGS. 

The construction of the spirometer bell is such that each millimeter 
length corresponds to a volume in the bell of 23c.c. ; hence by measuring 
the vertical distance between the bottom and top levels of the record 
made on the kymograph drum by the pointer at the beginning and end 
of every inspiration or expiration, the apparent volume of air inhaled 
or exhaled may be computed. By measuring all the rising portions 
of the respiration curve and subsequently multiplying the result by the 
known factor, the total ventilation of the lungs during the experimental 
period can be obtained. To simplify this calculation, a recording 
device has been added to the spirometer which is somewhat in the 
nature of a work-adder wheel 1 and permits the accumulative measure- 
ment of the movements of the spirometer bell in one direction. Each 
revolution of this wheel corresponds to a rise in the spirometer bell of a 
certain number of millimeters, and from the record of the number of 
revolutions of this wheel the apparent volume of air passing through 
the lungs can be calculated. 

In these experiments, the apparent volume obtained by this calcula- 
tion was converted to standard conditions of temperature and pressure 

by multiplying it by the fraction =^j in which p represents the baro- 

x Benedict, Deutsch. Archiv f. klin. Med., 1912, 102, p. 176. 



MECHANICS OF RESPIRATION. 161 

metric reading corrected for scale correction and diminished by 5 mm. 
This correction of 5 mm. was found desirable as a result of experiments 
in which the humidity of the air inside the spirometer bell was found to 
be usually about 30 per cent. As a matter of fact, calculations showed 
that the difference due to using an assumed value for complete satu- 
ration or partial saturation is not more than 1 or 2 per cent. In addi- 
tion to the correction for the pressure, the usual correction for tempera- 
ture was made. The total volume as reported is therefore the total 
ventilation per minute, corrected for 0° C. and 760 mm. and likewise 
for an average value of 5 mm., corresponding to the probable humidity 
of the air inside the spirometer bell. 

METHOD OF CALCULATING THE VOLUME PER INSPIRATION. 

The method of calculating the volume per inspiration is not so simple 
as it at first appears. Instead of simply dividing the total ventilation 
per minute by the number of respirations, most writers have been 
accustomed to calculating the volume per inspiration from the volume 
of the air converted to the conditions which exist in the lungs, that is, 
the prevailing atmospheric pressure less the tension of water-vapor at 
37° C. and corrected for the temperature of the lungs at 37° C. There 
has been considerable discussion, particularly in connection with the 
experiments of Galeotti, 1 and Loewy and Gerhartz, 2 as to whether the 
temperature conditions should be taken as 37° C, and whether the air 
is saturated at this temperature or not. This value is, however, most 
commonly used, and, indeed, we are not far in error in doing this, 
although, as was shown in an earlier publication, 3 the correct determi- 
nation of the temperature of the air in the lungs and the degree of 
saturation will obviously affect these computations somewhat. 

The method used for calculating our results is as follows : The total 
ventilation of the lungs, which has been reduced to standard conditions 
of 0° C. and 760 mm. pressure, is divided by the number of respirations. 
This value is then converted to the pressure existing in the lungs, which 
is the atmospheric pressure less the tension of aqueous vapor at 37° C, 
or 46.7 mm. It is subsequently converted to the temperature of the 
lungs by the usual calculation. A sample calculation will serve to 
show the method used : In the morning respiration experiment on April 
11, the ventilation of the lungs was 5.32 liters per minute at 0° C. and 
760 mm. The observed barometer was 758.7 mm. and the number of 
respirations per minute was 12.2. The volume per inspiration would 
therefore be 

760X (273+37) X5.32 



(758.7-46.7)X273Xl2.2 



= 529 c.c. 



'Galeotti, Biochem. Zeitschr., 1912, 46, p. 173. 

1 Loewy and Gerhartz, Biochem. Zeitschr., 1912, 47, p. 343. 

3 Benedict, Carnegie Inst. Wash. Pub. 77, 1907, p. 436. 



162 



A STUDY OF PROLONGED FASTING. 



RESULTS OF OBSERVATIONS ON THE MECHANICS OF RESPIRATION. 

The data secured by these methods regarding the respiration-rate, 
the ventilation of the lungs per minute, and the volume per inspiration 
give material for an interesting study of the effect of prolonged fasting 
upon the mechanics of respiration. These data are given in table 18, 
which shows two extensive series of values, one for the morning respi- 
ration experiments made directly after the subject came out of the 



Table 18. — Ventilation of lungs in experiments with L. at different times of the day, and with 
varying activity. (Respiration apparatus.) 



Date. 


Day of 
fast. 


Lying. 


Usually 8 h 30 m a.m. to 9 h 30 m a.m. T. 


rsually 7 p.m. to 7 h 45 m p.m. 


















Respira- 
tion-rate. 


Lung ven- 
tilation per 
minute. 1 


Volume -n 
per inspi- ^ 
ration.* 


spira- 
n-rate. 


Lung ven- 
tilation per 
minute. 1 


Volume 
per inspi- 
ration. 2 


1912. 






liters. 


C.C. 




liters. 


c.c. 


Apr. 11.... 




12.2 


5.32 


529 








12 




9.6 

9.6 

10.6 

9.3 


5.21 
5.19 
4.79 
4.97 


655 
650 
539 
639 
















13 




14 




15.... 


1st 


16.... 


2d 


10.9 


5.18 


576 
















17.... 


3d 


11.3 


5.24 


562 
















18.... 


4th 


9.8 


4.77 


591 
















19.... 


5th.... 


11.8 


4.88 


507 
















20. . . . 


6th 


12.0 


4.70 


473 
















21.... 


7th.... 


11.8 


4.79 


489 
















22.... 


8th.... 


10.7 


4.67 


530 
















23.... 


9th 


12.1 


4.65 


476 
















24. . . . 


10th 


10.9 


4.55 


504 
















25. . . . 


11th 


10.1 


4.40 


522 
















26.... 


12th 


12.8 


4.64 


429 1 


L2.8 


5^24 


488 


27.... 


13th 


12.8 


4.63 


437 ] 


14.9 


3 5.35 


"437 


28.... 


14th 


12.4 


4.61 


448 ] 


L4.7 


5.32 


437 


29. . . . 


15th 


12.3 


4.55 


446 ] 


L4.6 


5.83 


483 


30.... 


16th 


13.1 


5.00 


462 1 


L4.6 


6.01 


497 


May 1 


17th 


12.3 


4.81 


471 ] 


14.5 


5.79 


482 


2.... 


18th 


13.2 


4.61 


422 ] 


L5.1 


5.81 


465 


3.... 


19th 


12.8 


4.78 


449 1 


L4.7 


5.66 


465 


4.... 


20th 


14.3 


4.90 


413 








5.... 


21st 


10.0 


4.43 


532 ] 


L5!l 


5.76 


458 


6.... 


22d 


13.5 


4.91 


436 ] 


L4.9 


5.69 


460 


7.... 


23d 


14.0 


4.76 


410 ] 


L6.7 


6.03 


438 


8.... 


24th 


13.7 


4.69 


417 ] 


L5.5 


5.77 


456 


9.... 


25th 


14.2 


4.95 


428 ] 


L4.4 


5.74 


490 


10. . . . 


26th.... 


12.8 


4.75 


454 ] 


L3.6 


5.58 


502 


11.... 


27th.... 


12.8 


4.89 


461 ] 


L4.fi 


5.82 


483 


12.... 


28th 


14.8 


5.04 


410 ] 


L4.7 


5.76 


473 


13.... 


29th 


14.1 


4.96 


426 1 


L3.9 


5.72 


501 


14.... 


30th 


14.8 


4.80 


391 ] 


L3.9 


5.92 


514 


15.... 


31st 


13.3 


4.84 


438 








17 




9.9 
14.0 


3.93 
5.72 


485 
494 








18 























x The lung ventilation observed is here reduced to 0° C. and 760 mm. pressure. 
2 Calculated to the pressure existing in the lungs and to 37° C. 

3 During the period 3 h 16 m p.m. to 3 h 51 m p.m., with subject in lying position, the observa- 
tions were: respiration-rate, 13.4; lung ventilation, 5.14 liters; volume per inspiration, 466 c.c. 



MECHANICS OF RESPIRATION. 



163 



Table 18. — Ventilation of lungs in experiments with L. at different times of the day, and with 
varying activity. (Respiration apparatus.) — Continued. 



Date. 


Day of 
fast. 


Sitting. 1 


Period. 


Respira- 
tion-rate. 


Lung ven- 
tilation per 
minute. 2 


Volume 
per inspi- 
ration. 3 


1912. 

Apr. 16 

19.... 
23.... 
24.... 
26. . . . 
27. . . . 
29.... 

May 1 

4.... 

7.... 

14.... 


2d. . . . 

5th... 

9th... 
10th... 
12th... 
13th... 
15th... 
17th... 
20th... 
23d. . . . 
30th... 


4 h 00 m p.m. to 4 h 35 m p.m , , 
4 10 p.m. 4 43 p.m.* . . . 
3 52 p.m. 4 28 p.m... 

3 58 p.m. 4 57 p.m 

3 13 p.m. 4 11 p.m, .. 
12 14 p.m. 12 48 p.m 
3 23 p.m. 3 66 p.m.* . . . 
9 31 a.m. 10 04 a.m.*. . . 
9 35 a.m. 10 10 a.m.*. . . 

3 43 p.m. 4 14 p.m.* 

6 32 p.m. 7 02 p.m.*. .. 


10.3 
17.9 
16.7 
14.6 
15.8 
12.8 
18.7 
14.6 
15.3 
16.1 
17.8 


liters. 
5.58 
7.54 
5.48 
5.83 
5.37 
5.55 
7.88 
6.57 
6.22 
7.62 
8.05 


c.c. 

660 

517 

402 

484 

404 

525 

510 

542 

490 

573 

546 



1 Periods indicated by an asterisk were obtained with the subject sitting, writing. 
2 The lung ventilation observed ia here reduced to 0° C. and 760 mm. pressure. 
'Calculated to the pressure existing in the lungs and to 37° C. 

calorimeter and representing every day of the fast, and another series 
for the evening respiration experiments made each day during the 
latter part of the fast just before the subject entered the calorimeter. 

RESPIRATION-RATE. 

An examination of the respiration-rate for the morning period shows 
that there was a distinct tendency for it to increase as the fast con- 
tinued, the lowest rate being observed on the first day of the fast, i. e., 
9.3 respirations per minute and the highest rate of 14.8 respirations, on 
the twenty-eighth and thirtieth days of the fast. The values obtained 
in the evening began on the twelfth day of the fast and indicate a 
reasonably constant respiration-rate, averaging not far from 15 respi- 
rations per minute. The evening rate in practically all cases was 
slightly higher than the morning rate. 

Cathcart states that with his subject there was no change in the 
character of the respiration as the fast progressed, and his figures show 
a tendency for the morning respiration-rate to remain constant or to 
decrease slightly. Our observations, on the contrary, indicate a tend- 
ency to increase in the morning. Our findings also differ from those 
recorded by Luciani for Succi, as his curve indicates a tendency to fall 
towards the end of the fast. While on a number of days in Succi's fast 
the evening respiration-rate was higher than that obtained in the 
morning, in a large number of instances the reverse was true. In view 
of Succi's excitable temperament and the fact that his daily routine was 
not absolutely constant, it is more than likely that the discrepancies 
appearing between our results and Luciani's may be easily explained 
by the fact that in the experiment with Levanzin the routine was 



164 A STUDY OP PROLONGED FASTING. 

rigidly adhered to each day, the subject lying very quietly for some time 
while the respiration-rates were being recorded. 

The experiments of the Berlin investigators with Cetti and Breit- 
haupt were not sufficiently long to make them comparable with this 
31-day experiment and they were likewise complicated considerably 
by the fact that the subjects suffered from cold and colic. 

From an examination of all of the kymograph records obtained 
with L., it is clear that while prolonged fasting tended to increase 
the average respiration-rate, there was great regularity of respiration 
throughout each 15-minute period. Occasionally a deep breath was 
taken, but there was nothing like the great irregularity noted by Zuntz 
and his co-workers on the two Berlin fasters, an irregularity which may 
again be explained by the complications of cold and colic. 

VENTILATION OF THE LUNGS PER MINUTE. 

The actual amount of air passing through the lungs was measured on 
the spirometer and its recording attachment. The ventilation of the 
lungs per minute, which is given for each experiment in table 18, 
followed a somewhat singular course. In the morning observations 
the ventilation per minute showed a persistent, though slight, tendency 
to decrease during the first 4 days with food; it then rose perceptibly 
in the first 3 days of the fast and subsequently decreased until the low 
value of 4.4 liters was reached on the eleventh day. This was closely 
approximated on the twenty-first day, when a value of 4.43 liters was 
obtained. The lowest value in the experimental period was found on 
the second day with food after the fast, when the ventilation was 3.93 
liters. Few deductions can be drawn from these figures for the lung 
ventilation per minute, save that on certain fasting days the values 
were very low as compared with the four days preceding the fast, 
although, as has already been pointed out, the minimum value was 
obtained on the second day with food after the fast. Here again the 
values for the evening observations show an increase, the ventilation 
being invariably greater than during the morning experiments, rising 
at times as high as 6 liters. The average value was 1 liter higher than 
those obtained during the morning observations. 

VOLUME PER INSPIRATION. 

In discussing the values for the volume per inspiration given in table 
18, it must again be stated that these were not obtained by dividing 
the total ventilation of the lungs by the number of respirations, but 
by using the volumes changed to the conditions in the lungs, as is com- 
monly done by other writers of the present day. These figures show 
that there is a distinct tendency for the volume per inspiration to de- 
crease as the fast progressed, although certain high values are found on 
the eighth, eleventh, and twenty-first days of fasting. On the other 



MECHANICS OF RESPIRATION. 165 

hand, the lowest value recorded in the morning experiments — 391 c.c. — 
was on the thirtieth day of fasting. In the evening series we note that 
while the values in general are somewhat higher than in the morning, 
this increase seems to become greater toward the end of the fast. Thus, 
on the thirtieth day of the fast it was 391 c.c. in the morning and 514 c.c. 
in the evening. While, therefore, there is a positive average difference, 
inasmuch as in the evening the volume per inspiration is greater than 
in the morning, the difference has a tendency to become very much 
greater in the last week of the fast. 

INFLUENCE OF CHANGES IN BODY POSITION. 

On a number of days the subject was studied when sitting in his 
chair, either resting or writing. The values obtained are given in table 
18 for comparison with those found while the subject was lying on a 
couch. During the sitting experiments, when the subject was not 
writing, the respiration-rate was in practically all cases slightly higher 
than the values obtained in the morning respiration experiments while 
the subject was lying quietly. On the ninth day of the fast it increased 
from 12.1 to 16.7 respirations per minute. The ventilation of the 
lungs per minute also increased perceptibly in every instance, the in- 
crease being not far from 0.8 liter. On the other hand, the volume per 
inspiration varied considerably. In two instances there was a per- 
ceptible increase, on two other days it decreased, while on another day it 
remained essentially constant. This difference is not so apparent when 
the results are compared with the records for the evening respiration 
experiments. Unfortunately the sitting experiments were not suffi- 
ciently extended to draw any definite conclusions regarding the effect 
of the change in body position; furthermore, the whole study lacks 
suitable normal values for comparison. 

INFLUENCE OF THE WORK OF WRITING. 

On 6 of the fasting days the metabolism of the subject was studied 
while he sat in a chair and wrote actively. On 2 of these days he was 
studied in the forenoon and on 4 days in the afternoon. Since there 
is a tendency towards a diurnal variation in the mechanics of respiration 
between morning and evening, as shown by the increase in the respira- 
tion-rate and the ventilation of the lungs per minute, and the tendency 
for the volume per inspiration to increase, it is necessary to take this 
fact into consideration in discussing the results. In the two forenoon 
experiments there was in both instances an increase in the respiration- 
rate, a marked increase in the ventilation of the lungs per minute, and 
a great increase in the volume per inspiration. Inasmuch as the writing 
was accompanied by distinct, though perhaps slight, muscular effort, 
these findings are only what would be expected. In the afternoon 
experiments there was an increase in the respiration-rate much more 



166 A STUDY OF PROLONGED FASTING. 

noticeable than in the experiments in the forenoon. The ventilation 
of the lungs per minute showed a large increase, the values averaging 
about 7.75 liters per minute. There was also usually a measurable 
increase in the volume per inspiration. 

From these results it can be inferred that the slight muscular work of 
writing letters perceptibly affected the mechanics of ventilation in that 
the respiration-rate was somewhat increased and the ventilation of the 
lungs per minute noticeably so. So far as we know, no study has been 
made with normal individuals in which the ventilation of the lungs per 
minute and the volume per inspiration were so carefully observed as 
were those of our fasting subject, and hence we have no comparable 
values which will show to what extent the factors affecting the mechanics 
of respiration were influenced by prolonged fasting. It is reasonable 
to suppose, however, that muscular exercise of any kind would require 
a greater effort in the later stages of inanition. It is of particular 
interest that the lung ventilation per minute in the afternoon experi- 
ments was perceptibly greater than when essentially the same amount 
of work was carried out in the forenoon. 

INFLUENCE OF BREATHING AN OXYGEN-RICH ATMOSPHERE. 

On three days during the fast an experiment was made directly after 
the morning respiration experiment, in which the subject breathed an 
atmosphere containing from 95 to 75 per cent of oxygen. The influence 
of this increased amount of oxygen was distinctly noticeable with the 
ventilation of the lungs per minute and the volume per inspiration, 
although the respiration-rate changed but little. On the first day on 
which these experiments were made (May 12, 1912) the volume per 
inspiration with normal air was 410 c.c. and with the oxygen-rich 
mixture it was 487 c.c, an increase of 16 per cent. The values obtained 
in these experiments were as follows : Twenty-eighth day of fast, respi- 
ration-rate, 13.6, lung ventilation, 5.50, volume per inspiration, 487; 
twenty-ninth day, respiration-rate, 14.0; lung ventilation, 5.44, and 
volume per inspiration, 471 ; on the thirtieth day, respiration-rate, 14.2; 
lung ventilation, 5.34; volume per inspiration, 454. It is clear, there- 
fore, that with this subject the breathing of oxygen-rich mixtures 
resulted in a considerable increase in the ventilation of the lungs per 
minute, and while the respiration-rate was not materially affected, 
there was a considerable increase in the volume per inspiration. 

MAXIMUM EXPIRATION OF THE LUNGS. 

As an index of a possible change in the volume of the lungs and par- 
ticularly in the strength of the chest muscles, observations in regard to 
the maximum expiration of the lungs were made by Mr. Carpenter on 
5 days during the fasting period. For these observations a long rubber 
tube was attached to a 10-liter Bohr meter. The subject stood up and, 



MECHANICS OF RESPIRATION. 



167 



while holding his nose, inhaled as deeply as he could, then placed the 
end of the rubber tube in his mouth and exhaled into the meter to the 
smallest possible volume of the lungs. The difference between the 
beginning and end readings on the meter gave the maximum apparent 
expiration. In computing the true volume of the expiration, these 
figures were corrected for the temperature of the air in the gas-meter, 
the barometer, and the temperature of the air in the lungs, the latter 
being assumed to be 37° C. The results are given in table 19. 



Table 19. — Maximum expiration of subject L 


/. during fasting. 


Date. 


Day of fast. 


Time of 
observation. 


Volume 
observed. 


Barometric 
pressure. 


Volume 

exhaled 

(computed). 1 


1912. 
Apr. 15 

20 

May 7 

8 

14, , 


2d 

7th 

24th 

25th 

31st 


4 h 35 m p.m. 

3 00 

3 04 

3 30 
/4 30 
\4 36 


liters. 
3.45 
3.60 
2.90 
3.00 
2.50 
2.45 


mm. 

764.2 

762.0 

759.4 

752.6 

761.6 

761.6 


liters. 
3.74 
3.91 
3.15 
3.24 
2.71 
2.66 



1 In computing these values it was assumed that the temperature of the air when exhaled 
was 37° C, and when passing through the meter it was between 23° and 23.8° C. 

Although the volume on the seventh day of the fast (3.91 liters) was 
higher than that on the second day, a distinct tendency is shown for the 
volumes to decrease as the fast progressed. On the twenty-fourth and 
twenty-fifth days the volumes were essentially the same. Six days 
later — on the thirty-first day of the fast — the volume again materially 
decreased, as duplicate readings show values of 2.71 and 2.66 liters 
respectively. 

One can hardly ascribe this marked loss in the volume of air expired, 
amounting to about 30 per cent, exclusively to change in the volume of 
the lungs as a result of the fasting or exclusively to the inability of the 
weakened muscles of the chest to compress the chest walls further. In 
all probability both factors contributed to this change in the volume of 
the expiration. The readiness with which the lungs respond to arti- 
ficial atmospheric conditions leads one to believe that there may have 
been an absolute diminution in the available lung volume during the 
fasting period. On the other hand, there was unquestionably a falling 
off in the strength and general tone of the subject and he may not have 
been able to compress the lungs sufficiently to force out a large volume 
of air at the end of the fast. 



ALVEOLAR AIR. 
By Harold L. Higoins. 

Observations were made of the carbon-dioxide percentage of the 
alveolar air nearly every day throughout this fasting experiment. This 
offered an index as to the acidity of the blood and also an opportunity 
to study the control and mechanics of respiration throughout the fast. 

Alveolar air is the air which is in or comes from the alveoli of the 
lungs. As the active exchange of carbon dioxide and oxygen between 
the blood and the lungs takes place in the alveoli, it is readily seen that 
the tension or partial pressures of the different gases in the alveoli 
(carbon dioxide, oxygen, and also nitrogen, argon, etc.) will be very 
nearly the same in the alveoli as in the blood leaving the lungs. Inas- 
much as the quantity of a gas dissolved in a liquid is proportional to 
the partial pressure, and not to the percentage of the gas, the compo- 
sition of alveolar air is therefore probably better expressed in tensions 
or partial pressures than in percentages. 

SIGNIFICANCE OF ALVEOLAR AIR. 

Haldane and Priestley 1 have shown that carbon dioxide is the pre- 
vailing stimulus to respiration under normal conditions. Thus, if the 
carbon-dioxide tension in the respiratory center falls below a certain 
level, apncea is the result; and if, on the other hand, it rises above this 
level, the respiration volume is greatly increased and hyperpncea sets 
in. In other words, the respiratory center by respiratory impulses 
automatically keeps its carbon-dioxide tension constant. But the 
carbon-dioxide tension of the respiratory center is largely controlled by 
that of the arterial blood and the latter is, as mentioned previously, 
essentially that of the alveolar air. Haldane has therefore introduced 
the use of alveolar carbon-dioxide tension and shown that in any one 
individual it is practically constant under normal conditions, although 
the normal values of individuals may differ markedly from each 
other. It has been discovered that when there is an increased 
acidity of the blood, as in diabetic acidosis, or with reduced barometric 
pressure, as in high altitudes, the alveolar carbon-dioxide tension is 
lower than normal, and a smaller tension of carbon dioxide stimulates 
respiration. This has led to the presentation of the theory, 2 now 
quite satisfactorily established, that the H-ion concentration of the 
blood rather than the carbon-dioxide tension is the predominating 
factor in the control of respiration. Thus, when the H-ion concentra- 
tion (or degree of acidity) of the blood coming to the respiratory center 
reaches a certain level, impulses are sent out from the center to increase 



haldane and Priestley, Joum. Physiol., 1905, 32, p. 225. 
2 Winterstein, Archiv f. die ges. Physiol., 1911, 138, p. 167. 



168 



ALVEOLAR AIR. 169 

the respiration so that the net result is always the same H-ion concen- 
tration in the center. 

The acidity of the blood may be divided into two parts, that due to 
carbon dioxide and that due to other acids. As the total acidity 
necessary to cause respiration must always be the same, it is readily 
seen that if the other acids in the blood increase in amount, less carbon 
dioxide is necessary to raise the acidity to the point of stimulation of the 
respiratory center. Thus, one may say that the quantity of carbon 
dioxide will vary conversely from that of the other acids of the arterial 
blood. Since alveolar carbon-dioxide tension represents so closely the 
carbon-dioxide tension of the arterial blood, it affords a very good 
index of the acidity of the blood. It was mainly for this reason that 
the alveolar carbon dioxide in the experiment with L. was so closely 
followed. In fact, it seems that this index of the degree of acidosis is 
much more satisfactory and important than the urinary tests for acidity 

NH — N 
(as /3-oxybutyric acid, — i — » total titratable acidity, etc.), because 

the former represents the acid actually in the blood, while the latter 
only represents that excreted from the body. The other factors which 
affect the alveolar carbon-dioxide tension, such as the absorption of 
food and varying postures, were avoided with L., and thus one is able to 
study the results almost purely from the point of view of blood acidity. 

METHODS OF DETERMINING THE ALVEOLAR AIR. 
HALDANE METHOD. 

Haldane's method 1 for determining the alveolar carbon-dioxide ten- 
sion is the oldest and probably theoretically the most sound of any of 
the methods now in use. By it one collects two samples of alveolar air 
from different phases of the respiratory cycle and averages their carbon- 
dioxide content. The two phases chosen are immediately at the end 
of an inspiration, which is approximately when the alveolar carbon- 
dioxide tension is lowest, and at the end of an expiration, when the 
alveolar carbon-dioxide tension is nearly at its highest point. The 
subject breathes normally for some time; then at the end of a normal 
inspiration he makes a rapid, deep expiration through a tube about 
2 cm. in diameter and about 150 cm. long, sealing with his tongue the 
end he has just breathed into. A sample of the air in the tube near 
the mouth is then taken. This sample is considered to be alveolar air, 
as the air in the dead space of the respiratory passages and in that part 
of the tube from which the sample is taken has previously been pushed 
out by the air from the alveoli. Similarly a sample is taken of air 
forced through the tube from the lungs at the end of a normal expira- 
tion. Instead of sealing off the end of the tube with the tongue, use 

^aldane and Priestley, Journ. Physiol., 1905, 32, p. 225. 



170 A STUDY OF PROLONGED FASTING. 

has been made in our laboratory of a simple Siebeck 1 valve, which puts 
the subject under much less strain, as he does not have to hold his tongue 
to the tube while the sample is being taken into a gas-sampler. The 
average of the two analyses gives very closely the composition of the 
alveolar air. 

The Haldane method requires considerable attention on the part of 
the subject and, as it was feared that possibly in the course of the long 
fast the subject would not be physically able to co-operate very satis- 
factorily, the method used in these tests was modified somewhat. In 
view of what we now know of the condition of the subject throughout 
the fast, we may feel assured that this method would have been very 
successful ; but as several samples are often required to be sure of good 
agreement, and as it was probable that the subject's time would be 
much occupied, it was decided to modify the method somewhat to be 
sure of better agreement on fewer samples. 

It has been observed that, in the Haldane method, holding the breath 
for several seconds before the expiration does not cause the percentag ; 
of carbon dioxide in the alveolar air to increase with very great rapiditye 
this is naturally to be expected, for as the carbon-dioxide tensions of 
the alveolar air and the blood coming to the lungs approach the same 
figure, the increase in the former is slower. Furthermore, it appears 
that if a subject has previously been breathing somewhat abnormally 
for not over three or four respirations, the percentage of carbon dioxide 
in the alveolar air, after holding the breath for a few seconds, will be 
nearer that of the alveolar air when the breath is held similarly after 
normal respiration than is the percentage of carbon dioxide in the 
alveolar air of the same two cases when the breath is not held. Thus, 
it would seem that small deviations from the normal, such as appear in 
conscious respiration, would not be disturbing to agreeing results and 
that in a very small number of determinations (seldom more than two) 
figures can be obtained which are very good duplicates and which will 
bear a constant relation to the true alveolar carbon-dioxide tension 
when the subject is in the same position (sitting quietly). 

The modified method used in these tests, which later is called the Hal- 
dane method, is as follows: The subject began by breathing normally 
into the room through a short (5 cm.) tube connected with the Siebeck 
valve. Then at the end of an inspiration, selected by the observer who 
was watching the respiration, the subject was told to hold his breath. 
At the end of 5 seconds, timed by the observer, during which the valve 
had been opened, the subject breathed out rapidly and deeply through 
the long tube as in the Haldane method. After the expiration the valve 
was again closed. Usually two samples were taken each day in which 
the subject held his breath 5 seconds and two in which he held his breath 
8 seconds. The results obtained when the subject held his breath 8 

x Dr. R. Siebeck, of Heidelberg, has devised an ingenious slide-valve for this purpose, which 
may be secured of Universitats Mechaniker Runge in Heidelberg. 



ALVEOLAR AIR. 171 

seconds average a trifle higher than when the breath was held for 5 
seconds, but the agreement is so close that one could not satisfactorily 
select the individual determinations of each class if the results were put 
together and not labeled. The results were averaged, therefore, with- 
out distinction as to time. The determinations made in duplicate by 
this method agreed in general to 1 part in 20 and usually closer. To 
determine how close the results were to the figures which would have 
been obtained by the Haldane method, we experimented by both 
methods on ten different subjects sitting; the average result when the 
breath was held 5 seconds was 8.3 per cent higher and when held 8 
seconds was 9.4 per cent higher than with the Haldane method. The 
averages of the 5 second samples and the 8 second samples are thus 
about 9 per cent higher than the Haldane figures. Excluding two of 
the ten cases (5 per cent and 19 per cent), none showed differences of 
more than 3 per cent (6 per cent to 12 per cent) from the average 
difference (9 per cent). Thus, for comparing the daily observations 
with each other, it appears that the values obtained with the 5-second 
and 8-second methods in the experiment with L. are practically as 
significant as if the Haldane method were used. 

PLESCH METHOD. 

Use was also made of the Plesch method 1 applied by Porges, Leim- 
dorfer and Markovici 2 to clinical cases. By further modification of 
the method I have been able to get very constant duplicates with a 
minimum amount of attention by the subject. The apparatus used in 
this method consists of a woman's rubber bathing cap (pure gum), 
which is fastened to the bottom of an inverted shallow copper pan 
(about 20 cm. in diameter). On the other side of the pan is soldered a 
f-inch (2 cm.) three-way valve; by means of a rubber- tube connection 
the subject may breathe back and forth through this valve, either from 
the room or from the bag made of the bathing cap and the pan. A 
small brass stop-cock is attached to the pan, from which a sample of the 
gas in the bag may be obtained. In a determination, the bag was first 
emptied and 600 c.c. of the room air was admitted, the measurement 
being made by a meter. The subject then began breathing room air 
through the rubber tube and three-way valve, closing the nose with 
the thumb and forefinger of the hand holding the apparatus. At the 
end of a normal expiration the observer turned the valve and the 
subject breathed in all of the 600 c.c. of air in the bag. He then 
breathed back and forth at the rate of one complete respiration in 
5 seconds, the time being followed by the observer, who instructed 
the subject when to breathe in and when to breathe out. At the end of 
4 complete respirations, %. e., 20 seconds, the three-way valve was 
turned and a sample taken in the gas-analysis apparatus for analysis. 5 

Plesch, Zeitschr. f. exp. Path. u. Therapie, 1909, 6, p. 380. 
2 Porges, Leimdorfer and Markovici, Zeitschr. f. klin. Med., 1911, 73, p. 389. 
3 AU the analyses of alveolar air were made on a portable Haldane apparatus. (Haldane, Methods 
of Air Analysis, London, 1912.) 



172 A STUDY OF PROLONGED FASTING. 

This method is probably the most adaptable for use with the average 
patient, when the condition of acidosis is being compared from day to 
day, or when a gross picture of the degree of acidosis is desired. The 
method does not give the carbon-dioxide tension of the arterial blood, 
but seems rather to approach the carbon-dioxide tension of the venous 
blood, because, as the same air is rebreathed, it is obvious that the 
alveolar air, the arterial blood, and venous blood will all have eventu- 
ally the same carbon-dioxide tension, namely, that of the venous blood, 
because it is the highest. For this reason, especially as it is the normal 
carbon-dioxide tension of the arterial blood, which is the important 
factor in the regulation of the respiration, this method theoretically is 
not so important as the Haldane method. 1 But with the subject in 
the same position and with the same amount of previous activity, we 
have found that the carbon-dioxide tension determined by this means 
bears a very constant relation to that of the Haldane method; this was 
assured from numerous comparisons of the different methods on many 
normal individuals, the results being about 20 per cent higher than the 
values obtained with the Haldane method. The same relation may 
also be observed with L., as the results in table 20 show that, excepting 
on the first few days, the difference between the 5- and 8-second Haldane 
method and the modified Plesch method is about 10 per cent. 

METHOD OF CALCULATING ALVEOLAR AIR FROM RESPIRATION 
EXPERIMENTS. 

The morning and evening respiration experiments, which were made 
with the universal respiration apparatus, included the determinations 
of the carbon-dioxide production, oxygen consumption, respiratory 
quotient, pulse- and respiration-rates, and inspiratory ventilation of 
the lungs. These experiments also give some data regarding the alve- 
olar carbon-dioxide tension and the dead space of breathing, which are 
of interest in considering the other alveolar-air determinations. The 
dead space in respiration is the air that is inspired and again expired 
without entering the alveoli, in which active gaseous exchange takes 
place, and thus is unchanged. The following formula for calculating 
the percentage of carbon dioxide in the alveolar air is therefore readily 
understood : 

A1 . ^~ C0 2 V = Total volume of air ex- 

Alv. per cent C02 = - — —-- — — • 7 • 

V-(DSXR) piredmcx. 

C0 2 = C0 2 production in c.c. DS = Dead space in c.c. 

R = Number of respirations. 

Naturally these factors must be measured for the same unit of time 
and under the same conditions of pressure, temperature, and aqueous 
tension. The unit of time chosen in the experiments with L. has been 

*The same criticism applies to the 8-second and 5-second Haldane method as used in these 
experiments, although probably to a less degree. 



ALVEOLAR AIR. 173 

1 minute, while the gas volumes considered in the application of the 
above formula have, for the sake of simplicity in calculation, been taken 
at 20° C, 760 mm., and dry. 

Elaborating this general formula for use in connection with and cal- 
culation from the respiration experiment, we get the following: 

A1 + rn CO,X 1.075 

Alv.p.ct.C0 2 = . p TpTT ~r\ _nr\ Z 

v(m) -°- 015 v (t1>) " ™<r L T Bt ) ~ < R x m 

C0 2 = C0 2 production in c.c. per minute at 0° C. and 760 mm. ; 
and C0 2 X 1.075 = C0 2 production at 20° C, 760 mm. 
2 = 2 consumption in c.c. per minute, 0° C, 760 mm. 
V = inspiratory ventilation of lungs per minute at barom- 
eter, temperature, and humidity prevailing in spirom- 
eter of respiration apparatus (i. e., about 20°C. and 
33 per cent or 66 per cent humidity). 

v(^) -0.015 V(^) - 1.075( ° 2 ~ C ° 2 ) -expiratory ven- 
tilation of lungs per minute at 760 mm., 20° C, dry. 
DS = dead space in c.c. (20° C, dry, 760 mm.) 
R = respiration-rate per minute. 
P = barometric pressure. 
1.075 = factor to convert gas volumes from 0° C. to 20° C. 
In a respiration experiment the inspiratory ventilation is obtained at 
the pressure, temperature, and humidity of the air in the spirometer on 
the apparatus. The temperature is not taken in each experiment, but 
as it is probably very close to 20° C, this temperature is assumed in 

P 

each calculation. The term =^r obviously reduces the ventilation to 

/ P \ 
760 mm. The term —0.015 V (;™) corrects for moisture in the 

spirometer; this moisture comes from two sources, namely, from the 
lungs of the subject and from the moistener used to prevent the air in 
the respiration apparatus from becoming too dry for comfortable respi- 
ration. Two kinds of moisteners were used during the fasting experi- 
ment. In the first part of the series (until April 22) a moistener con- 
structed of the lower part of a Kipp gas-generator was employed. In 
this the air before coming to the nosepieces bubbles through water; 
the humidity with this form of moistener has been found to be 66 per 
cent saturated; 1.5 per cent of the recorded ventilation was therefore 
water- vapor and accordingly subtracted. The other moistener was a 
piece of moist cheese-cloth in the tube leading to the nosepiece. When 
this was used, the humidity in the spirometer was found to be only 33 
per cent, so that in these experiments 0.75 per cent of the recorded 
volume was subtracted and not 1.5 per cent. The expiratory volume 



174 A STUDY OF PROLONGED FASTING. 

is smaller than the inspiratory volume, because the amount of oxygen 
consumed is greater than the amount of carbon dioxide produced. 
Accordingly we subtract 1.075 (0 2 — C0 2 ) from the inspiratory venti- 
lation. But during inspirations in a respiration experiment one-half 
of the carbon dioxide produced is absorbed in the soda-lime bottles, 
so that the recorded inspiratory volume is correspondingly increased; 
on the other hand a volume of oxygen equal in volume to one-half the 
volume consumed is added to the respiration apparatus during an 
inspiration and thus makes the recorded inspiratory volume corre- 
spondingly too small. Thus, instead of 1.075 (0 2 — C0 2 ), the factor 
for changing the inspiratory volume to the expiratory volume becomes 

1.076 (9q£% 

The only other factors in the equation are Alv. p. ct. C0 2 and DS. 
With either one known, the other may easily be determined. The 
formula for calculating the dead space is as follows : 

R 
When the experiments on L. were made, we had not considered the 
possible use of this equation and so have not the complete data for 
calculating either of these factors, but we still have sufficient material 
to draw some interesting conclusions. 

Assuming the personal dead space of breathing for the subject L., 
together with that of the nosepieces, etc., to be 120 c.c, the percentage 
of carbon dioxide in the alveolar air has been calculated for all of the 
morning and evening respiration experiments, and the results for each 
series have been averaged as shown in table 20. Also, making use of 
the alveolar-air figures found on the same day, the respiratory dead 
space in each experiment has similarly been calculated and averaged. 
As the alveolar air was not taken at the time of the respiration experi- 
ments and as the dead space might possibly have changed in size 
during the fast, fixed differentiation of the results is difficult. These 
results will be discussed later. 

CONDITIONS OF TAKING ALVEOLAR-AIR SAMPLES 

The samples were taken by the Haldane and Plesch methods with 
the subject sitting in an armchair. After the fast had begun, they were 
taken at about l h 35 m p.m. to 2 p.m. Between the taking of the samples, 
each of which was analyzed before another was taken, the subject was 
sitting quietly and usually reading. On several days he had visitors 
while the experiment was in progress. On one of these days, April 22, 
while talking with a visitor, he became quite excited. On the other 
days there was no marked excitement while the samples were being 
taken. On April 22, it is interesting to note that the alveolar carbon- 
dioxide tension by the Haldane method was very low — in fact, much 



ALVEOLAR AIR. 175 

lower than that calculated from the respiration experiments on the 
same day would seem to indicate it should normally have been. On 
2 days, April 25 and 26, the alveolar air was not determined. 

On the food days preliminary to the fast, the alveolar air was sampled 
with the subject sitting, immediately after the respiration experiment 
of the morning and the taking of the body-weight. On the morning 
of the first food day, April 11, the subject was unused to the apparatus 
and the tests were unduly hastened ; the results obtained can not there- 
fore be considered so reliable as on later days. The samples of alveolar 
air in the food days subsequent to the fast were taken at approximately 
l h 35 m to 2 p. m., as during the fast. 

DISCUSSION OF RESULTS. 

The results of the determinations made by the Plesch and Haldane 
methods are expressed in table 20, columns f and g, as tensions (milli- 
meters of mercury). The tensions are calculated from the carbon 
dioxide obtained by the analyses. From the prevailing barometric 
pressure is subtracted the figure 46.7 mm., which is the aqueous tension 
of air saturated at 37° C. (the air in the alveoli being saturated with 
water-vapor at this temperature), and the resulting pressure is multi- 
plied by the percentage of carbon dioxide found. Before discussing 
from a physiological point of view the results obtained by these methods 
it seems desirable to summarize first the results gathered from the 
respiration experiments. 

SIZE OF DEAD SPACE IN FASTING. 

A diminution in the size of the heart, liver, and other organs, as well 
as in the size of the muscular tissue, having been observed during the 
fast, the question was raised by Dr. Benedict as to whether or not the 
dead space in breathing also changed in size during the fast, and at his 
suggestion use was made of the formula given previously, and the data 
available, to calculate so far as possible a figure for the size of the dead 
space for each morning and evening experiment. To get a value for 
the alveolar carbon-dioxide percentage to use in these calculations, cer- 
tain corrections have been made in the values obtained by the 5-second 
and 8-second methods given in column g of table 20. These corrections 
are necessary for two reasons : first, because the alveolar air was taken 
with the subject sitting, while in the respiration experiments the sub- 
ject was lying on his back; second, because the results obtained by the 
Haldane method give a carbon-dioxide percentage about 9 per cent 
of the total less than that when the subject held his breath from 5 to 8 
seconds. As shown in a recent paper, 1 the alveolar carbon dioxide 
lying is about 106 per cent of that sitting. Thus the alveolar percent- 
age of carbon dioxide which should be used in calculating the dead 
space in the respiration experiments is 97 per cent of the figure from 
which column g is calculated, and is given in column a. In using 

Wiggins, Am. Journ. Physiol., 1914, 34, p. 114. 



176 



A STUDY OF PROLONGED FASTING. 



this value for each day, it was necessary to assume that the alveolar 
percentage of carbon dioxide had not changed between 8 a. m. or 
8 p. m., when the respiration experiments were made, and 2 p. m. of 
the same day, when the alveolar percentage of carbon dioxide was 
taken. It would seem, however, that if there were a change in the 





Table 20. — Alveolar-air and dead-space determinations in 


experiment with L. 










Day 


Alveolar air. 


Alveolar 
CO, 


Computed alveolar COj (dead 


space = 120 c.c ). 












Date. 


of 


(Haldane), 


From morning respiration 


From evening respiration 




fast. 


corrected 
to lying 
position. 1 


experiments. 




experiments. 




By periods. 


Average. 


By periods. 




Average. 






A 


B 




C 


D 




£ 


1912. 




p. ct. 


p.ct. 


p.ct. 


p.ct. 


p.ct. 


mm. Hg. 


p.ct. 


p.ct. 1 p.ct. 


p. a 


. mm.Hg. 


Apr. 11 






4.76 




5.11 


4.94 


35.2 




.... 








12 




4.89 


4.70 


4.84 


4.75 


4.76 


34.0 


















13 




4.94 


4.80 


4.89 


4.99 


4.89 


35.1 


















14 




5.10 


4.94 


5.25 


5.12 


5.10 


36.9 


















IS 


1st 


4.43 


4.76 


4.91 


4.60 


4.76 


34.3 


















16 


2d 


4.29 


4.68 


4.63 


4.42 


4.58 


32.6 


.... 
















17 


3d 


4.37 


4.29 


4.30 


4.30 


4.30 


30.6 


















18 


4th 


4.36 


4.42 


4.35 


4.37 


4.38 


31.1 


















19 


6th 


4.33 


4.50 


4.65 


*4.43 


4.47 


31.5 


















20 


6th 


4.30 


4.39 


4.51 


4.43 


4.44 


31.7 


















21 


7th 


4.38 


4.49 


4.43 


4.48 


4.47 


32.0 


















22 


8th 


3.94 


4.36 


4.31 


4.36 


4.34 


30.9 


















23 


9th 


4.45 


4.44 


4.30 


4.32 


4.35 


30.3 


















24 


10th 


4.29 


4.23 


4.31 


4.21 


4.25 


30.4 


















25 


11th 




4.43 


4.23 


4.19 


4.28 


30.8 


















26 


12th 




4.42 


4.42 


4.27 


4.37 


31.9 


3.65 


3.64 


3.63 


3.64 


[ 26.4 


27 


13th 


3.88 


4.43 


4.33 


4.40 


4.39 


31.4 


3.71 


3.64 


3.68 


3.6* 


\ 26.1 


28 


14th 


3.95 


4.17 


4.10 


4.15 


4.14 


29.6 


3.92 


3.58 


3.45 


3.61 


26.1 


29 


15th 


3.92 


4.27 


4.14 


4.04 


4.15 


29.7 


3.29 


3.21 


3.21 


3.24 


23.1 


30 


16th 


3.74 


3.79 


3.75 


3.76 


3.77 


26.9 


3.04 


3.05 


3.07 


3. OS 


21.8 


May 1 


17th 


3.87 


3.82 


3.81 


3.74 


3.79 


27.2 


3.03 


3.14 


3.14 


3.1C 


22.2 


2 


18th 


3.90 


4.10 


3.87 


3.83 


3.93 


28.1 


3.08 


3.07 


3.15 


3.1C 


22.2 


3 


19th 


3.75 


3.87 


3.83 


3.72 


3.81 


27.3 


3.13 


3.11 


3.15 


3.13 


22.4 


4 


20th 


3.64 


3.82 


3.75 


3.67 


3.75 


26.9 












5 


21st 


3.87 


3.92 


3.79 


3.92 


3.88 


27.9 


3.03 


3.05 


3!o4 


3.04 


21.9 


6 


22d 


3.77 


3.74 


3.63 


3.54 


3.64 


26.2 


3.07 


2.96 


3.21 


3.08 


22.1 


7 


23d 


3.78 


3.71 


3.95 


3.75 


3.80 


27.2 


3.12 


2.90 




3.01 


21.4 


8 


24th 


3.68 


3.97 


3.80 


3.86 


3.88 


27.5 


3.12 


2.98 


3! 14 


3.08 


21.7 


9 


25th 


3.77 


3.79 


3.74 


3.60 


3.71 


26.1 


2.95 


2.97 


3.00 


2.97 


20.9 


10 


26th 


3.84 


3.74 


3.76 


3.92 


3.81 


26.9 


3.11 


3.14 


3.15 


3.13 


22.2 


11 


27th 


3.72 


3.64 


3.81 




3.73 


26.7 


2.92 


3.01 


3.03 


2.99 


21.4 


12 


28th 


3.78 


3.78 


3.71 


3.48 


3.66 


26.2 


2.92 


2.96 


3.00 


2.96 


21.1 


13 


29th 


3.84 


3.73 


3.62 


3.65 


3.67 


26.1 


2.91 


2.95 


2.96 


2.94 


20.8 


14 


30th 


3.77 


3.83 


3.87 


3.69 


3.80 


27.2 


2.89 


2.88 




2.89 


20.7 


15 


31st 


3 4.33 


3.60 


3.64 


3.47 


3.57 


25.6 












16 




3 4.36 






















17 




3 4.83 


4.73 


4.68 


2 4.'77 


4.70 


33^2 












18 






4.06 


4.15 


2 4.06 


4.11 


29.3 













1 Obtained by taking 97 per cent of the percentages alveolar CO2 from which the figures in column a 
■were calculated. 

Calculated percentages for the fourth period on April 19, 4.31; May 17, 4.60; May 18, 4.18. 

3 The subject ended his fast with the taking of fruit juices and honey on the morning of May 15, afte 
the conclusion of the respiration experiments. 



ALVEOLAR AIR. 



177 



alveolar air during that time, it would be proportional on each day; 
on certain days subsequently specified there is good reason to believe 
that the change was not proportional. The results of these calcula- 
tions are given in column h for the morning experiments and j for the 
evening experiments, with the averages in columns i and k, respectively. 

Table 20. — Alveolar-air and dead-space determinations in experiment with L. — Continued. 



Alveolar air. 


Volume of dead space (using alveolar CO2 in column A). 






COj tension for 
















sitting. 


Computed from morning 
respiration experiments. 


Computed from evening 


Day 








respiration experiments 


. 


of 




Modifica- 


Haldane 














Date. 


tion of 


method 














fast. 


















Plesch 
method. 


(breath held 
5"to 8"). 


By periods 


1. 


Aver- 
age. 


By periods. 


Ave 
age 


r- 




F 


G 


H 




I 


J 




K 






mm. Hg. 


mm. Hg. 


c.c. 


c.c. 


c.c. 


c.c. 


c.c. 


c.c. 


c.c. 


c.c 




1912. 


31.9 


31.7 






















Apr. 11 


35.4 


36.0 


141 


124 


132 


132 
















12 


35.1 


36.5 


135 


124 


116 


125 
















13 


35.7 


37.5 


130 


110 


118 


119 
















14 


33.7 


32.8 


85 


80 


98 


88 














1st 


15 


33.7 


31.3 


90 


84 


108 


94 














2d 


16 


33.1 


32.1 


127 


126 


126 


126 














3d 


17 


35.4 


31.9 


115 


121 


120 


119 














4th 


18 


34.1 


31.4 


109 


96 


^ll 


109 














5th 


19 


34.7 


31.6 


114 


105 


111 


110 














6th 


20 


35.3 


32.3 


112 


117 


113 


114 














7th 


21 


34.4 


28.7 


86 


86 


81 


84 














8th 


22 


34.3 


32.1 


120 


130 


128 


126 














9th 


23 


34.0 


31.5 


126 


118 


126 


123 














10th 
11th 
12th 


24 
25 
26 


32^8 


28 A 


84 


88 


82 


85 1 


32 


136 ] 


133 ] 


34 


[ 13th 


27 


32.4 


29.1 


103 


109 


107 


106 3 


22 


145 ] 


56 ] 


41 


14th 


28 


31.6 


28.9 


96 


105 


111 


104 ] 


69 


175 1 


L79 ] 


74 


L 15th 


29 


30.8 


27.5 


116 


119 


118 


118 ] 


78 


184 ] 


L77 ] 


8C 


1 16th 


30 


30.0 


28.5 


124 


125 


130 


126 1 


90 


176 ] 


L79 1 


85 


17th 


May 1 


31.4 


28.7 


107 


122 


125 


118 1 


83 


179 ] 


179 1 


8C 


> 18th 


2 


29.3 


27.6 


111 


114 


122 


116 1 


68 


172 ] 


66 ] 


OS 


19th 


3 


30.1 


26.9 


107 


112 


118 


112 










20th 


4 


30.0 


28.7 


115 


127 


116 


119 1 


79 


187 ] 


82 1 


83 


21st 


5 


30.5 


27.9 


122 


130 


138 


130 1 


76 


181 ] 


64 1 


74 


22d 


6 


29.5 


27.8 


125 


110 


122 


119 1 


65 


185 


1 


75 


23d 


7 


29.1 


26.8 


100 


112 


108 


107 1 


64 


175 1 


60 1 


6C 


24th 


8 


29.7 


27.3 


119 


122 


131 


124 1 


89 


185 1 


82 1 


8fi 


25th 


9 


29.2 


28.1 


127 


126 


114 


122 1 


85 


179 1 


77 1 


8C 


26th 


10 


29.4 


27.5 


126 


113 




120 1 


91 


178 1 


75 1 


81 


27th 


11 


29.9 


27.9 


120 


125 


140 


128 1 


90 


184 ] 


83 1 


86 


28th 


12 


29.1 


28.1 


127 


135 


133 


132 1 


97 


196 ] 


93 1 


95 


29th 


13 


29.7 


27.8 


116 


114 


125 


118 2 


08 


192 


2 


0G 


30th 


14 


2 33.2 


2 31.8 


















31st 


15 


2 35.0 


2 32.0 




















16 


2 38.0 


2 35.1 


126 


129 


*124 


128 '. 












17 
18 



'Calculated volume of dead space for the fourth period on April 19, 121 c.c; May 18, 133 c.c. 
2 The subject ended his fast with the taking of fruit juices and honey on the morning of May 15, after 
tJ e conclusion of the respiration experiments. 



178 A STUDY OF PROLONGED FASTING. 

Considering each series by itself, the conclusion may be drawn from 
both the morning and the evening experiments that there is no constant 
change in the size of the dead space of breathing as a result of the fast. 
Although there is more or less fluctuation from day to day, yet the 
general average of dead-space volumes found at the beginning of the 
fast is much the same as that toward the end of the fast. A number 
of very low values may doubtless be explained quite well as follows: 
On April 15 and 16 there is a markedly lower value for the dead space; 
this is probably due to an especially large drop in the alveolar air 
during the day, as one might perhaps expect, these two days being the 
first two of the fast. As previously mentioned, the subject was much 
excited on April 22, when the Haldane samples were taken, and as a 
result the value obtained for the alveolar air was probably low; thus 
the calculation makes the dead-space figures also too low. On April 27 
and 28 there is also an indication of a lower dead space; this is like- 
wise probably due to a change in the alveolar air, as it will be noted that 
table 20 shows a marked fall in the alveolar carbon-dioxide tension 
about this date. 

DIFFERENCE IN MECHANICS OF RESPIRATION IN MORNING AND EVENING. 

On comparing the values for the dead space calculated from the 
evening experiments with those for the morning experiments, one finds 
a constantly higher dead space, which is, on the average, 55 c.c. This, 
of course, is based on the assumption that the alveolar percentage of 
carbon dioxide is the same in the morning as it is at night. Since some 
physiologists believe that the dead space is always essentially the same, 
it seems desirable to consider how large a difference in the alveolar 
percentage of carbon dioxide must have existed between the morning 
and evening experiments to indicate so marked a change in the venti- 
lation. Assuming for the size of the dead space the figure 120 c.c, 
which represents approximately the mean value previously calculated 
for the dead space in the morning experiments, the alveolar air has 
been computed for each respiration experiment, as shown in columns 
b and d and the averages in columns c and e. The alveolar percentage 
of carbon dioxide (carbon-dioxide tension) in the morning experiments 
shows, in general, the same changes that the alveolar carbon-dioxide 
tensions by the other methods have indicated. In fact, the alveolar 
carbon-dioxide tensions obtained in this manner very satisfactorily 
supply the values for April 25 and 26, when the alveolar air was not 
taken by the other methods. The close agreement of the values for 
the alveolar percentage of carbon dioxide in the several duplicate experi- 
ments gives evidence of the even and normal respiration of the subject 
and the great care in making the respiration experiments. 

If the alveolar carbon-dioxide tensions of the morning and evening 
experiments of the same day are compared, an average difference is 



ALVEOLAR AIR. 179 

found of 5.4 mm. (0.7 per cent). It is difficult to interpret accurately 
this change in the mechanics of ventilation, but it is clear that it exists, 
for with a given output of carbon dioxide, the respiration volume is 
much larger in the evening than in the morning. Two causes for this 
are possible — one, a lower percentage of carbon dioxide in the alve- 
olar air, the other an increased volume in the dead space of respira- 
tion. Both causes may be in part responsible for the difference. If 
the alveolar air had been taken in the evening and in the morning at the 
same time as the respiration experiments, the exact cause could have 
been located; but unfortunately the data are not available. However, 
this change seems of sufficient importance to summarize possibilities. 

A lower alveolar carbon-dioxide tension in the evening will mean, 
perhaps, a higher acidity of the blood toward evening, or possibly a 
respiratory center more sensitive to a given stimulus. FitzGerald and 
Haldane 1 have noted that the alveolar carbon-dioxide tension falls as a 
subject becomes mentally tired. Ordinarily this fall would not be 
noticed during the day, as the food eaten tends to raise the alveolar 
carbon-dioxide tension and thus renders the figures uncertain; but in 
a one-day fasting experiment with myself as subject, I failed to find 
any fall in the alveolar carbon-dioxide tension up to 4 h 30 m p. m., when 
the experiment stopped. 2 Changes in the dead space have been reported 
by Douglas and Haldane, 3 who state that with muscular work a 
larger dead space is found, which would lead to the general conclusion 
that the dead space increases with increasing metabolism. Krogh, 4 
however, using different experimental methods, maintains that the 
dead space is practically always the same. As the metabolism in the 
evening experiments is only about 10 per cent higher than in the morn- 
ing, it seems unlikely that this higher metabolism would of itself cause 
the change in the dead space, especially as the subject was at complete 
rest and in the same position in both cases. In experiments carried 
out by Dr. J. H. Means, of the Massachusetts General Hospital, and 
myself, we have found changes in the dead space as the result of drugs 
which correspond very closely to such changes as may have occurred 
here. It seems quite possible that the bronchi became dilated in the 
latter part of the day, having lost some of their tone with increasing 
fatigue. Any changes in the dead space which may have occurred 
in the experiments with L. are of such size that they can readily be 
explained by dilation of the bronchi. 5 

In this connection it may be well to state that the dead space given 
in table 20 includes the volume of the inspired air, measured at 760 mm. 
and 20° C, dry, which does not reach the alveoli, where active exchange 

FitzGerald and Haldane, Journ. Physiol., 1905, 32, p. 486. 
2 Higgins, Am. Journ. Physiol., 1914, 34, p. 116. 
•Douglas and Haldane, Journ. Physiol., 1912, 45, p. 235. 
4 Krogh, Journ. Physiol., 1913, 47, p. 30. 
B Siebeck, Skand. Archiv f. Physiol., 1911, 25, p. 81. 



180 A STUDY OF PROLONGED FASTING. 

of gas takes place. As L. was breathing through the three-way valve 
and nosepiece connected to the apparatus, the figures given are 30 c.c. 
higher than his actual personal dead space, since the subject with 
each respiration drew from the air system of the apparatus 30 c.c. 
which did not reach either the respiratory passages or the lungs. 

SIGNIFICANCE OF CHANGE IN THE ALVEOLAR AIR DURING THE FAST. 

Finally, it is advisable to compare the data of the alveolar carbon- 
dioxide tensions as the fast progressed. Columns c, f, and g, in table 
20, serve this purpose the best. As one would naturally expect, there is a 
drop in the alveolar carbon-dioxide tension with the increased acidosis 
of the fast. The subject ate the last meal before fasting on the evening 
of April 13; on the afternoon of April 15 the fall in the alveolar carbon- 
dioxide tension is first noted. This fall is about 4 to 5 mm. (2 mm. 
Plesch method) ; after this the alveolar air continues at about the same 
level until April 27 or 28, when a second quite sharp drop, also of about 
4 to 5 mm., is apparent. The new level is maintained with only slight 
fluctuations until the end of the fast. With the taking of food again, 
there is a rise in the alveolar carbon-dioxide tension, as would be 
expected with the resulting diminution of acidosis. On the morning 
of May 18 a slight fall in the alveolar carbon-dioxide tension is again 
noted in connection with the respiration experiments before breakfast. 
Possibly a part of the rise on May 15 and subsequent days with the 
Plesch and Haldane methods may have been due not only to the dimi- 
nution of acidosis, but to an effect similar to that since noticed in this 
laboratory with the ingestion of food. 1 

In connection with the second sharp fall in the alveolar carbon- 
dioxide tension, it is of interest to note several parallel experimental 
findings. The chlorine excretion in the urine on April 27 and 28 
dropped from a previously high level to a lower figure, at which it 
continued for the remainder of the fast. A rise in the total volume of 
the urine occurred also at about this time. The daily nitrogen excre- 
tion in the urine for the first 10 or 12 days of the fast was slightly 
over 10 grams; there was then a fall to about 8 grams, which was 
maintained throughout the rest of the fast. It may be noted that this 
drop in the nitrogen excretion is simultaneous with the drop in the 
alveolar carbon-dioxide tension. 

Since alveolar air is intimately associated with the acidosis of the 
subject, one naturally looks for simultaneous changes in the factors of 
the urine which are taken as indicators of acidosis. Thus the /3-oxy- 
butyric acid seems to show a rise to a high level about the twelfth day 
of the fast; such a change is difficult to judge, however, as the /3-oxy- 
butyric-acid variations from day to day are quite large. The other 
index, the ratio of ammonia nitrogen to total nitrogen, also shows a 

Wiggins, Am. Journ., Physiol., 1914, 34, p. 117. 



ALVEOLAR AIR. 181 

simultaneous rise by reason of the increase of the ammonia and the 
decrease of the total nitrogen. 

From these data one may safely conclude that there is a marked 
increase in the acidity of the blood in the fast, beginning on the second 
day; the acidity then did not change markedly until about the four- 
teenth day of the fast, when another decided increase in blood acidity 
occurred. The recovery to normal acidity in the blood begins to be 
evident in the first few days after the fast. 

CONCLUSIONS. 

The results for the alveolar air and dead space may be summarized 
as follows: 

(1) On the second day of the fast, the carbon-dioxide tension in the 
alveolar air showed a drop from the normal value. It remained at this 
new level until about the fourteenth day of the fast, when there was a 
second rather sharp fall, after which no further marked change occurred. 
Each of these falls is about 4 mm. Thus the blood acidity may be said 
to have markedly increased on the second day of the fast and to 
have remained at this higher level until the fourteenth day, when a 
second increase occurred; there was no further change until the end 
of the fast. 

(2) There is no sign of an accumulative change in the size of the dead 
space from day to day as the fast progressed. 

(3) A change in the mechanics of respiration on the respiration 
apparatus between morning and evening experiments during the fast 
shows that there was either a marked change in the alveolar air or 
else a change in the size of the dead space during the course of each 
day. If the former, the alveolar air fell about 6 mm. between morn- 
ing and evening, returning during the course of the night to essen- 
tially the morning figure. If the dead space changed, it increased in 
size about 55 c.c. and again became normal by the next morning. 



SUBJECTIVE IMPRESSIONS AND MENTAL ATTITUDE TOWARD 

THE FAST. 

By Harry W. Good all, M. D. 

The freedom of speech characterizing this subject, his preconceived 
ideas on fasting and on the humanitarian service of his fast led to excep- 
tionally full comments on the whole project and specifically his subjec- 
tive impressions. His habit of thought and introspection probably 
make them of average value, though admittedly they are recorded not 
as scientific observations, but as general indices to his mental makeup, 
his personal experiences and his beliefs, as outlined freely to the writer 
on each visit. 

SUBJECTIVE IMPRESSIONS. 

April 14, 1912 (18 hours after beginning the fast) : 

The subject states that he is very happy in the thought that the fast has 
actually begun. The value of the experiment to the world can not 
be estimated, and after a few days the mind will be clear and active. 
In explaining the influence of fasting upon the mind, he stated that 
on his long journey from Malta he had been obliged to eat food that 
was poisonous, more especially animal foods, and that his body was 
saturated with this poisonous waste, making the mind dull and causing 
a kind of physical fatigue. "When these waste matters are elimi- 
nated the mind will be clear, and I will feel buoyant and hopeful." 
In referring to the nocturnal emission which occurred during the 
night of April 12-13, he stated that one of the most important things 
noted in connection with his fasts was the behavior of the sexual 
organs. During fasting there is a reversion to the animal type, a 
periodicity of sexual desire at monthly intervals. In speaking of his 
subjective sensations since beginning the fast, he states that he feels 
perfectly well, has had no sensations of hunger, and no thought of 
food. He has been a mouth breather for years. Is troubled with 
naso-pharyngitis and tinitus aurium. Both these conditions always 
improve with fasting. 
April 16, 1912 (third day of fast) : 

Feels perfectly well. Has had no sensation of hunger, and no abdominal 
sensations, aside from slight rumbling of gas in the intestines. Has 
passed very little odorless gas by rectum; there has been no belching 
of gas, no sense of fatigue. Mind is not yet clear enough for active 
mental work. 
April 18, 1912 (fifth day of fast) : 

Feels perfectly well. No sensation of hunger. No longing for food, but 
occasionally thinks of the agreeable taste of ice cream. No sense of 
muscular weakness or fatigue. Has passed a little odorless gas by 
rectum. Naso-pharyngitis and tinitus better. 
April 20, 1912 (seventh day of fast) : 

Mentally depressed yesterday and to-day. He attributes this to the rain 
and cloudy weather, as he always feels depressed when the sun does 
not shine. Feels as well physically as usual. Expressed his satis- 
faction at the manner in which the fast was being conducted. 

182 



SUBJECTIVE IMPRESSIONS AND MENTAL ATTITUDE. 183 

April 22, 1912 (ninth day of fast) : 

Feels well and hopeful again. Only complaint is the sensation of a dry 
coating of the pharynx and a bad taste in the mouth. No sensation of 
hunger. Not conscious of his stomach. Has passed a little odorless 
gas. 
April 24, 1912 (eleventh day of fast) : 

Is conscious of slight muscular weakness but otherwise feels well. No 
loss of ambition. No sensation of hunger and reading about food 
does not stimulate a desire to eat. There is no dryness in the pharynx 
today. Still passes a little odorless gas by rectum. Says he is thirsty 
for the first time. 
April 26, 1912 (thirteenth day of fast) : 

Still conscious of slight muscular weakness. Has no special inclination 
for mental work. Still has the sensation of thirst. No sensation of 
hunger. Still passes a little odorless gas by rectum. Expressed 
satisfaction at the progress of the fast. 
April 28, 1912 (fifteenth day of fast) : 
No depression to-day. Mind is clear now. Desires to study. No sen- 
sation of hunger. Mouth, which has tasted bad since the ninth day 
of the fast, is now improving. Passes very little odorless gas by 
rectum. Had nocturnal emission at 6 a. m. 
April 30, 1912 (seventeenth day of fast) : 
Mind is clear. Feels well. Ambitious to study. Conscious of slight 
muscular weakness. No sense of hunger. 
May 2, 1912 (nineteenth day of fast) : 

Feels somewhat weaker physically, but mind is clearer and can do better 
mental work. Thinks the poisons of the food ingested previous to 
the fast are about eliminated now. Complains of a "bilious taste" 
in the mouth to-day. No sensation of hunger. 
May 4, 1912 (twenty-first day of fast) : 

Feels practically the same as he did May 2, except that he is a little 
depressed by the cloudy weather and by remaining indoors. 
May 6,1912 (twenty-third day of fast) : 

Feels a little brighter to-day. Conscious of physical weakness. Mouth 
does not taste so bad. No desire for food. 
May 8, 1912 (twenty-fifth day of fast) : 
Feels very well. No desire for food. Still passes a little odorless gas by 
rectum. 
May 10, 1912 (twenty-seventh day of fast) : 

No change from last note. Still has a bad taste in his mouth. 
May 12,1912 (twenty-ninth day of fast): 

Sense of physical weakness, but to-day is very ambitious for his studies 
and writing. Feels more hopeful and clearer mentally. Pleased 
with the progress of the experiment. No desire for food. 
May 14, 1912 (thirty-first day of fast) : 

Depressed mentally because he has to break his fast tomorrow. States 
that the fast should not be broken until the tongue has become clean 
and a desire for food has returned. This in his opinion would take 
several more days. To break the fast at this time is injurious. 
Aside from this he is pleased with the manner in which the fast has 
been conducted, and congratulates himself that he has been able to 
go through with it successfully. Says the time has passed very 
quickly. To-day he feels well. Has no particular weakness, but has 
been conscious of some muscular fatigue since the eleventh day of 



184 A STUDY OF PROLONGED FASTING. 

May 14, 1912— Continued. 

the fast. This, however, is no more marked than it is many days 
when he is taking regular meals. The most pronounced physical 
change is a sensation that his body is very light. This has gradually 
developed as his weight has decreased, and necessitates his measuring 
his steps when he walks. Says that at no time has he felt like 
reclining on account of body fatigue. His neurasthenia is greatly 
improved, and during the entire fast has only shown itself as slight 
despondency and some irritability on the rainy days. His mind has 
been much clearer throughout the fast than it is when food is taken. 
To-day his mind is clear. He has better imagination. He is full of 
hope and courage. Is ambitious to do mental work. Has had no 
sensation of hunger, no sensations of faintness. He had no desire 
for food, aside from the pleasant thought of ice cream on the third 
and fourth days of the fast, and has been glad that he did not have 
to eat. Has had no abdominal pain or discomfort. His naso- 
pharyngitis is much improved, and his tinitus has practically dis- 
appeared. 
Maylld, 1912 (two hours after breaking fast) : 

Extremely depressed and despondent because he had been obliged to break 
his fast before his body was prepared for it. Felt very weak physi- 
cally. The foods that he selected in breaking his fast were concen- 
trated orange and lemon juice, grape juice, and honey. Experience 
had taught him that these were the only natural, rational foods to be 
taken at this time. Meat broths and other animal foods were poi- 
sonous. In his opinion these natural foods should be followed first 
by cooked fruits, then vegetables, and later a return to the ordinary 
diet. Stated that he had no appetite, and that nothing tasted good, 
although the lemon juice was not unpleasant. 

10 a. m. Food was first ingested. This immediately caused a sensation 
of warmth in the stomach, and he was conscious of a pulsation in the 
epigastrium. There was no desire to belch gas, no nausea or other 
symptoms until — 

ll h 45 m a. m. when he began to have distress in the abdomen, starting in 
the epigastrium and radiating towards the right hypochondrium. 
The pain was dull, intermittent in character, but not colicky. He 
describes it as a sensation of distension of the stomach accompanied 
by rumbling and belching of gas. Furthermore, he believed he could 
feel the progress of the distension of the alimentary tract as the 
ingested food moved along the intestines, and he felt by — 

12 u 10 m p. m. that the food had proceeded as far as the right iliac fossa. 
Then at— 

1 p. m. the rumbling and belching of gas was much more marked and the 
pain was somewhat more severe. Marked lassitude and depression. 
Soon after pain became easier, until — 

4 h 15 m p. m., when the character of the pains changed from the "pains of 
distension" to intermittent attacks of cramp-like or colicky pain, 
which became very severe. These were experienced about the um- 
bilicus and in the lower abdomen. Between the attacks he was very 
drowsy. The colic gradually increased in severity until — 

5 p. m., when he defecated for the first time. This was accomplished 
without difficulty or pain. There was some gas, and the movement 
had a very bad odor. After this he was free from pain until — 

5 h 45 m p. m., when the cramps began again, increasing in severity, the 
location being the same. This time the pain made him extremely 



SUBJECTIVE IMPRESSIONS AND MENTAL ATTITUDE. 185 

May 15, 1912— Continued. 

weak. There was profuse perspiration and intense thirst. This 
continued until — 

7 h 15 m p. m., when he had a second movement of the bowels, liquid in 
character. After this he was nauseated, and the pain continued 
severe until — 

9 h 4& m V- m -> when he vomited and then began to feel better, and at — 

11 p. m. the pain stopped. 
May 16, 1912 (second day after breaking fast) : 

About 2 a. m. began to have severe abdominal cramps again, continuing 
until — 

2 h 30 m a. m., when he had a third movement of the bowels, liquid in 
character. After this he was comfortable, but passed gas at frequent 
intervals. The odor of the gas was very disagreeable, but the passage 
was not accompanied by pain. Was comfortable until — 

8 p. m., when he stated that he was in no physical distress, but that he 
was very despondent. He appeared hysterical, crying a good deal. 
He was offended because he had been obliged to break the fast, and 
complained that this, together with a few disagreeable experiences 
with some of the men in charge of the experiment, and the lack of 
fresh air and exercise throughout the fast, was responsible for his 
physical and mental weakness. He attributed the abdominal pain 
to the presence of gas. Inasmuch as he had not defecated during the 
fast, "a hard plug of intestinal secretion had accumulated in the 
rectum, and when he took food the advancing bolus compressed the 
gas present in the intestines, causing pain." 
May 17, 1912 (third day after breaking fast) : 

Feels very weak, but no abdominal pain for past 24 hours. Passed a 
fairly comfortable night, but did not sleep as well as usual. Is still 
depressed and emotional, but to a much less degree since the pain 
stopped. Is taking the same kinds of food, but well diluted with 
water, as was suggested to him. Feels much better physically, 
but has no ambition for mental work. Late yesterday afternoon 
he was very thirsty for about an hour, drinking 5 glasses of water. 
Up to the present time he has eaten the juice of two lemons diluted 
with equal parts of water and sweetened with a teaspoonful of honey, 
also 1,500 c.c. of orange juice diluted one-third with water, to which 
enough honey was added to make it sweet. Says his general condi- 
tion is not nearly so good as it was on the last day of the fast. His 
appetite is just beginning to return, but he is not hungry yet. 
May 18,1912 (fourth day after breaking fast) : 

Extreme mental depression. Marked excitability, hysterical in character. 
Weeps when spoken to, and his voice is scarcely audible. Complains 
of general weakness, almost prostration. Sensation of trembling all 
over the body. Did not sleep one moment during the entire night. 
Beginning at 5 a. m. he had three loose movements of the bowels 
preceded by severe colicky pains. He attributes this diarrhea to 
worry. Can not bear the thought of food to-day. Explains his 
mental state as being due to his general dissatisfaction at his treat- 
ment at the laboratory. Later in the day, at his own request, he 
was sent to a private room in the Massachusetts General Hospital. 
Upon admission his condition was fully described by the writer to 
the resident physician and the visiting physician. 



186 A STUDY OF PROLONGED FASTING. 

May 19, 1912 (fifth day after breaking fast) : 

Was seen at the Massachusetts General Hospital at 9 a. m. At the time 
he was sitting on one of the verandas reading. While he still appeared 
somewhat emotional, he said he was no longer depressed, but was 
in a most cheerful state of mind. He expressed his regret for the 
trouble he had caused at the laboratory, and said he was so nervous 
and irritable that he did not realize just what he was doing. He sent 
his apologies to Professor Benedict, and wished me to say that he 
would be glad to go back to the laboratory and go into the calorimeter 
again, if it would add anything to the value of the experiment. 
When asked about his comfort at the hospital, he said he was very 
much pleased with the care and attention he was getting. His 
appetite had returned, and he had relished his breakfast of "Boston 
baked beans," which he ate for the first time. 
May 20, 1912 (sixth day after breaking fast) : 

The hospital reported that he had voluntarily left the institution. His 
reason for leaving could not be learned, but in so far as could be 
determined he had no grievance. 
October 19, 1912 (five months after breaking fast) : 

On account of the sensational stories which were circulated, an effort was 
made to obtain as much information as possible regarding the personal 
experiences of the subject after leaving the hospital. He was first 
questioned as to the reason for leaving the hospital so abruptly. 
First of all he stated that he was not sincere in his remark regarding 
the hospital, as noted under the date of May 19. His reasons for 
misrepresenting the true condition was a fear that he would not get 
good treatment if he made any complaints. He was quite dissatisfied 
at the hospital. In the first place, his pride was injured in that he 
was not given the attention which was due an individual of his stand- 
ing. One grievance was the fact that he was recorded on the chart 
as "A laborer from the island of Java." Another was the fact that, 
although he was in a room by himself, there were a good many rooms 
in the ward and he was obliged to use the common toilet and bath- 
room. Still another was the fact that the physicians at the hospital 
had no experience in fasting and did not order the proper food for 
him. The day following his admission he was given milk and eggs, 
and he believed that these animal foods would poison his blood and 
spoil all the benefits of the fast. The ingestion of the food was 
followed by cramps and diarrhea, and he thought he was delirious. 
The following morning (May 20) he looked at his chart, and finding 
his temperature recorded at 99 degrees and his pulse-rate 90, he 
knew he was being poisoned and hastily left the institution. 

When he reached the street he did not know where to go, inasmuch as he 
was entirely unfamiliar with Boston. Being a newspaper editor 
himself, he thought he would find trustworthy advice as to where to 
go by consulting an editor of one of the Boston papers. He then 
took a cab to the office of one of the newspapers, the editor of which 
engaged a room for him at one of the large hotels. In his room at 
the hotel he was visited by some of the staff of the paper and talked 
the situation over with them. He was surprised to find an account 
of his fast, with photographs of himself, in the paper the following 
morning, as he thought he was speaking in confidence and he was 
not aware that photographs were being taken. 



SUBJECTIVE IMPRESSIONS AND MENTAL ATTITUDE. 187 

October 19, 1912— Continued. 

After spending one night at the hotel, he was disinterestedly ( ! ) advised 
by the editor to go to some secluded place to avoid annoyance from 
reporters from other papers. In accordance with this advice the 
editor arranged for his care at Bridge water, Mass., agreeing to take 
care of his expenses for three weeks, and later start him on a lecture 
tour in return for information concerning the experiment and his 
treatment at the laboratory. He went to Bridgewater, accompanied 
by a reporter who visited him each day, and was delighted with the 
place. On the way down he talked freely with the reporter. The 
next morning he was given a copy of the paper, and upon reading the 
article which concerned himself, he became angry and excited, 
because he did not know the things he said were to be published, and 
declared that he never made some of the remarks pertaining to 
Professor Benedict. He told the reporter that his action was dis- 
honorable, and that if such reports were to be continued he would go 
to another paper with his story. It was then agreed that the paper 
should only publish what he himself wrote and signed. 

After this the paper was not sent to him, and it was 5 days before he could 
obtain the copies. To his surprise and anger he found they had not 
published the best part of his stories, and had put in things that he 
did not write and which were detrimental to those in charge of the 
experiment. He again protested, saying it was not honest treatment, 
and finally refused to have anything more to do with the paper. The 
newspaper defrayed his expenses for 9 days, and then he went to the 
house of a fellow countryman in East Boston. After this he gave 
some lectures, and later was taken up by the Esperanto Society. 

Later still a gentleman became interested in him, and offered to defray 
his expenses while studying medicine. At the present time, i. e., 
date of this examination October 19, 1912, he is a student at the 
Harvard Medical School. Each morning on his way to the School 
he passes Professor Benedict's window in an automobile, which, he 
said, with much satisfaction, is the " irony of the case." He concluded 
his remarks by saying he had encountered a good deal of trouble with 
various persons he had come in contact with since arriving in America, 
but he felt now that for the most part it was due to misunderstanding. 

MENTAL ATTITUDE OF THE SUBJECT TOWARD THE FAST. 

The mental attitude of the subject was noted at each visit during 
the entire period of observation. The predominating idea with him 
throughout was that fasting is always beneficial. He believed that 
it is followed by physical and mental well-being in normal persons 
and is the rational treatment for diseased conditions. His own expres- 
sions were: "Food impedes the body and mind and animal food is 
poison. * * * I am anticipating the fast with much pleasure, as 
the poison of the food I have eaten will be eliminated, my body cleaned 
of its impurities, and my mind will be free and active. * * * The 
neurasthenia and depression from which I have suffered for years will 
leave me and I shall feel free and light and full of hope." Any dissent 
from these ideas was promptly resented with such a remark as: "My 



188 A STUDY OF PROLONGED FASTING. 

dear doctor, you know nothing about fasting, while I have made it a 
scientific study." He considered himself an authority, qualifying the 
assertion by saying that he is now and always has been a student, that 
he has repeatedly fasted, carefully watching the effects of fasting, and 
that he has studied all the available writings upon the subject. He 
stated that his object in undertaking the present experiment was to 
accomplish the most complete fast yet undertaken, under the best 
scientific conditions possible to obtain, not for his own enlightenment, 
but to demonstrate to the world beyond doubt the truth of his theories. 
He said : " The experiment I am about to undertake will be of the great- 
est benefit to mankind." 

Upon careful questioning it was learned that he had never undertaken 
a fast under strict scientific observation and that his reading had been in 
the main confined to non-scientific works, largely magazine articles. 

On the occasion of the first visit it was evident that he was not only 
willing but anxious to assist in every way possible the work that was 
being done. He was very cheerful and was plainly pleased with atten- 
tion shown him. This cheerful attitude continued for the first week. 
He was interested in what was being done and apparently tried to 
describe his subjective feelings with exactness. If any attempt was 
made to oppose his ideas as to fasting, his usual smile disappeared 
quickly and he assumed a sober, slightly injured air. 

On the seventh day he appeared downcast, his movements were 
less active, and he was decidedly depressed mentally. He attributed 
this to the cloudy weather and rain and in so far as could be determined 
no other reason for the change existed. He expressed no displeasure 
at the manner in which the experiment was being conducted. On the 
ninth and eleventh days he was again cheerful, but not so enthusiastic 
as he had been during the first week, and he moved about as though 
he felt some physical fatigue. He admitted that he experienced 
muscular weakness. Any conversation, however, which was pleasing 
to him would arouse his enthusiasm for a short time. From this time 
up to the twenty-ninth day of the fast he was frequently depressed, 
but always courteous and ready to submit to the examination. On the 
days with sunshine he was always more cheerful, but at no time was he 
as enthusiastic as during the first week. He appeared to be slightly 
fatigued most of the time after the first week. His movements in 
preparing for the examinations were more deliberate, and any attempt 
to hurry him was politely resented. After the first week he gradually 
became more sensitive to discomfort and pain, complaining of any 
unusual pressure of the stethoscope or pressure of the hands in palpat- 
ing the organs. He frequently spoke of the annoyance of the rectal 
thermometer and of the adhesive tape used in retaining the stethoscope 
on the chest wall. This annoyance was plainly shown in the expression 
of his face and in the careful manner in which he moved about when the 



SUBJECTIVE IMPRESSIONS AND MENTAL ATTITUDE. 189 

thermometer was in the rectum. During the periods of depression he 
was frequently disinclined to talk, and was sometimes irritable. At no 
time, however, did he object to the examination, and he always seemed 
willing to do whatever was necessary for the success of the experiment. 

On the fifteenth day he said that his mind was beginning for the first 
time to become clear; on the nineteenth day he began to feel the desire 
to do mental work; and from this time on he declared that his mind was 
continually growing clearer. Certainly there was no outward evidence 
of the truth of this statement. He appeared fatigued mentally, and 
he neither understood nor answered the questions put to him so 
promptly as he did early in the fast. As far as could be determined 
there was no such stimulation for study as he had predicted. On the 
contrary he seemed to be less occupied with his books and papers. 

On the last day of the fast he appeared to be very sober and assumed 
rather an injured air. His general attitude was a complaining one. 
The fast was being broken contrary to his judgment, as it was harmful 
to take food before the desire for food had returned. Then for the first 
time he complained of his medical care, saying that while the examina- 
tions were made in the most careful and painstaking manner no atten- 
tion had been paid to his physical exercise and he had not been allowed 
to go out in the fresh air as much as he should have. 

Nothing was observed at any time which would lead one to suppose 
that the subject experienced any sensation of hunger or any feeling of 
distress in the abdomen throughout the entire fast. 1 

Two hours after breaking the fast he was seen seated at a table 
where he was slowly eating his fruit juice and honey. His expression 
was downcast and his features drawn. His voice was weak and he 
spoke with deliberation. Notwithstanding his resentment at the break- 
ing of his fast he willingly submitted to examination. He complained of 
pain in the abdomen and sudden spasmodic changes in his expression 
occurred at the time he said he was having pain. Palpation of his 
abdomen, however, only slightly intensified the discomfort. 

Twenty-four hours after breaking the fast the general expression of 
depression was more marked. The voice was weak and he moved 
about very slowly. At this time there was no evidence of any physical 
discomfort other than lassitude. 

On the third day after partaking of food he appeared in decidedly 
better spirits. There was no expression of discomfort and he smiled 
frequently during the examination. His movements were not so 
deliberate. He attributed his bad feelings of the previous days to the 
pain and discomfort he had suffered and did not appear to entertain 

'Cetti on the fifth day of the fast complained of pain in the epigastrium from time to time, on 
the sixth day of belching gas and of distress in the abdomen, on the eighth day of severe abdominal 
pain, which disappeared on the ninth day after the bowels moved ; on the tenth day he complained 
of feeling very weak physically and of being nauseated. No note is made of the gastro-enteric 
condition in the case of Breithaupt and Beaute. 



190 A STUDY OF PROLONGED FASTING. 

such a strong feeling of resentment because he had been obliged to 
break his fast. 

On the fourth day after taking food his mental attitude had entirely 
changed. He was very emotional, his voice scarcely audible. He 
wept as he talked. His hands trembled and his face was bathed in 
perspiration. He appeared weak physically, and while he made no 
objection to the examination, any undue haste or unusual pressure 
on the body made him complain. He evidently felt that he had not 
been given the proper medical attention during the entire period, 
although he had refused, at all times, to accept suggestion, except that 
he dilute his fruit juice. This was because "the physicians had no 
experience in fasting." He demanded that he be sent to a hospital, 
where he could get the attention he needed in his present sick, weakened 
condition. 

The following morning, when visited at the hospital, his entire 
attitude had again changed. He was sitting on a veranda reading, and 
appeared delighted at seeing me. He was in a very cheerful state of 
mind, but was still emotional. He moved about quickly, showing none 
of the prostration of the previous day. He appeared to be very sorry 
that he had been so unreasonable at the laboratory, was apologetic, 
and expressed his willingness to return. 

When seen five months after the fast was broken, he appeared rather 
unhappy. In telling his story it was evident that he had had differ- 
ences with nearly every person he had come in contact with. He was 
plainly disappointed because the world had not given him the recogni- 
tion due him for the sacrifice he had made for the benefit of mankind. 



THE PSYCHO-PHYSIOLOGY OF A FAST. 

By Herbert Sidney Langfeld, 
Instructor in Psychology, Harvard University. 

The subject of these tests was a man 40 years of age, of medium 
height and slender. When not in conversation his manner was languid 
and it is perhaps due partly to this that he seemed to lack physical 
strength and vigor. In temperament he was of the decidedly emotional 
southern type, sensitive, quick to anger, loquacious, credulous, and 
fertile in imagination. This last characteristic is probably responsible 
for the fact that the unusual appealed to him. Once having espoused 
a cause or entertained an idea, he would hold to it tenaciously. He was 
a man of a few fixed ideas or complexes, which formed the basis of his 
mental life. 1 

The tests herein described lasted from April 11 to May 15 inclusive. 2 
Food was taken on April 10, 11, 12, and 13 and again on May 15. 
The intervening 31 days were fast days. The psychological tests were 
made at 5 p. m. each day and lasted 1 hour. During the half hour 
before the tests the subject rested. 

The psychological tests were made under unusually accurate and 
complete control of diet and occupation. One factor important to 
mental measurement was found to vary, that is, the mood of the subject. 
As far as L.'s willingness to cooperate is concerned, there was nothing 
to indicate to the experimenter a change in this attitude or that his 
general interest in the work relaxed at any period of the series. On the 
other hand, there is no doubt that he was happier during the first part 
of the fast, rather depressed and silent in the middle, and somewhat 
irritable and excitable toward the end, although this irritation was at 
no time directed toward the tests. The greatest depression occurred 
after a prolonged continuation of bad weather and very much decreased 
after he was able to go out in the air. He was also much happier after 
having received visitors. He himself remarked that the monotony of 
the program was the most difficult thing he had to endure. As to 
his physical condition he made few complaints. He felt well through- 
out and insisted that he had no sense of hunger even during the first 
days. 3 The only discomfort of which he spoke was the coated con- 

1 An idea of his intelligence and interests may be obtained from the association reactions. 
See Appendix II, pp. 222-229 

2 The tests on April 1 1 were tentative and are not included in the curves. 

•This is contrary to the experience of most fasters. W. B. Cannon and A. L. Washburn (An 
Explanation of Hunger, Am. Journ. Physiol., 1912, 29, p. 441) describe the feeling of hunger aa 
follows: "Hunger . . . is a dull ache or gnawing sensation referred to the lower mid-chest 
region and the epigastrium. It is the organism's first strong demand for nutriment, and, not 
satisfied, is likely to grow into a highly uncomfortable pang, less definitely localized as it becomes 
more intense." Further (p. 442): "There is abundant evidence, however, . . . that during 
continued fasting hunger wholly disappears after the first few days." Professor Cannon has 
recently informed the author that from what certain fasters have told him he believes that sensa- 
tions of hunger may be absent from the beginning; that in fact some people may never have the 
sensations of hunger as just described. 191 



192 A STUDY OF PROLONGED FASTING. 

dition of his tongue and the unpleasant taste in his mouth. It was 
his idea that the fast should continue until this disappeared and it was 
for this reason that he was loath to break his fast on the thirty-first 
day. 1 Although he seemed more feeble toward the end of the fast and 
gave one the impression of a man convalescing from a weakening 
illness, yet he was always able to walk without assistance and at no 
time was it necessary to omit or alter a test through lack of strength on 
his part. On May 15, the day he broke his fast, he suffered severe 
colic, induced by the food he ate, and although tests were made the con- 
ditions were most unfavorable. It had been planned to continue the 
examination for several weeks longer, inasmuch as such tests would 
obviously be of inestimable value for comparison with the fasting tests. 
Unfortunately that was quite impossible under the circumstances and 
an entire year elapsed before further records could be obtained. 

Several factors influenced the selection of the tests. In the first 
place the time was limited. There was only 1 hour daily available and 
it seemed advisable to arrange for as many tests as possible during this 
hour in order to obtain a good mental picture. It was therefore neces- 
sary to choose short tests and also those requiring the minimum of 
effort, as one test had to follow the other without pause for recupera- 
tion. For example, prolonged tests for fatigue would have been of 
great value, but they could not be considered. In the second place, the 
fasting began a few days after L.'s arrival and little time could be 
devoted to preliminary trials in order to obtain the best combination, 
and the program once arranged could not be fundamentally changed. 2 
After consultation with Professor Raymond Dodge, a series of tests were 
selected. A few days' experience, however, showed the necessity of 
several alterations, and the revised program was as follows: (1) Rote 
memory for words; (2) tapping test; (3) strength test; (4) tactual-space 
threshold; (5) touch threshold; (6) free association and reproduction 
reactions; (7) association reactions, genus-species; (8) association reac- 
tions, noun- verb; (9) cancellation test; (10) hand- writing; 3 (11) visual 
acuity; (12) memory for words after 55 minutes. Later the touch 
threshold, which was taken on the under part of the lower forearm 
with a von Frey hair, was discontinued on account of the impos- 
sibility of obtaining reliable results in a short period of time. The 
association reaction genus-species was also omitted through difficulty 
in finding sufficient reaction words of equal simplicity. In addition to 
the tests L. was requested to describe all the dreams he had on the pre- 
vious night. 4 This was given before the visual acuity test. All the 

thirty days were considered sufficient for the physiological tests and he was allowed one day 
more to excel Succi's record. 

2 A few minor changes were introduced. 

3 A superficial examination of the daily records revealed no change. A systematic examination 
of the data has not yet been made. 

4 See Appendix I, p. 222. 



THE PSYCHO-PHYSIOLOGY OF A FAST. 



193 



tests with the exception of that of visual acuity were made in a small 
room free from disturbing influences. 1 

The general conditions of the experiments and the nature of the 
tests having been described, each test will now be treated separately, 
first as to the particular conditions and second as to the results. 

MEMORY FOR WORDS. 

Ten one-syllable words were chosen and these were read twice to 
the subject, who recalled as many as possible immediately after the 
the second reading. After 55 minutes the subject again attempted to 
recall these words. 



tVoof 



words 


















































































W 












































































































A 
























li- 


\ 


s 


\ 




/ 










tJS. 




































/ 








I 


\J 






\: 






















































\ 






1 


k 


1 










: 




























\ 


































\ 


i 


\ 


i 


\ 




\ 






\ 


1 








































\ 


! 


\ 


i 


























r T ,. , , \^ //rimed/ate ■ 
J. Memory for »ords\ , 






























• 


1 


' 






















































£ 






























































































































































































































































































































































" _, . . . , . .. (A Greatest No be fore f/rsr mistake- 
II. Memory for crtgirsX _ . . 

\B wttnon/y o/ic mistake - 




A/oof 








































digits 








































H 


















































































9 




/ 


s 














A 














: 


\\ 
















/ 


\ 




























8 




/ 








V 






/ 












/ 






*, 














\ 


( 




/ 




















\ 


/ 
















\ 


/ 


\ 


/ 








// 










\ 


/ 














\ 














^ 






\ 






\ 












/ 


\ 












/ 






























\ 


* 




\ 








1 






















/ 






\ 


/ 






\ 


IB 















































































































































I ays 123456 

t 
fxrsf began 



7 9 9 10 II li IS /* 15 IS 17 It 13 to II ti li 14 IS tS 17 U 29 30 31 32 33 3+ I t 

Fasf ended Later tests 
Fig. 23. — Memory testa. 

From the curves (figure 23) it will be seen that there are marked 
fluctuations, a circumstance which is always met with in mental tests 
and which will be found in all the curves. It will therefore be only 
possible to speak of general tendencies throughout. In the curve for 
immediate rote memory (A) it will be seen that the poor record made 
on the eleventh day (the third day of the test) only occurs once again, 
and that on the twenty-fourth day, while a perfect score of the 10 words 
was made 3 times and all of them during the last two-thirds of the fast. 
It can be said that although the early records reoccur frequently 
toward the end, yet the curve as a whole shows a slight general improve- 
ment, but so slight that not much significance can be attached to it. 

x It is much to be regretted that time and conditions prevented tests for the threshold of audition 
and smell. 



194 A STUDY OF PROLONGED FASTING. 

The curve B, indicating the amount of retention after 55 minutes, on 
the other hand, shows a more or less steady improvement until near 
the end of the series, and even when these last trials are included the 
general tendency of the curve is decidedly upward. In 4 instances, 
and these all in the last two-thirds of the series, the retention curve 
crosses the rote memory curve, which means that on these days the 
retention after the lapse of almost an hour was better than the im- 
mediate memory. L., upon being questioned, was emphatic in his 
assurance that he never thought of the words in the interim, so that 
this relative improvement in retention was not due to any conscious 
repetition during the pause. 

TAPPING TESTS. 

The instrument used was similar to the tapping-board described by 
Whipple. 1 It consisted of a board 12 cm. square and covered with 
aluminum. This metal is not very well adapted for the tapping-board, 
but it was selected for its lightness, it being thought quite probable that 
the tests would have to be made toward the end of the experiments 
with the subject lying down and the board resting on his chest. The 
stylus also had an aluminum point. The records were taken on a 
kymograph. The tapping lasted for 30 seconds and periods of 10 
seconds were marked off on the records. The subject being left-handed 
used that hand. As he was over-sensitive to cold during the fast he 
wore, besides a heavy woolen undershirt, a heavy dressing-gown, which 
added to the weight he had to lift. Neither the hand nor arm was 
allowed to rest on the table during the tapping. 

Curve III (figure 24) shows a gradual improvement for the first 
6 days, when the maximum of the series — 215 taps or about 7 taps 
per second — was reached. The curve then descends for the next 
9 days, when the minimum of 170 taps was reached. From this 
point to the end of the series there is a rise to a point just below the 
maximum. This rise is not, however, gradual, but consists rather of 2 
plateaus, one of 9 the other of 7 days, separated by decided jumps and 
followed by a gradual but very marked end spurt of 4 days. 

The initial improvement can well be due to practice in using those 
particular sets of muscles, combined with increasing familiarity with the 
work. This same rise also occurred in the dynamometer tests. The 
drop, however, begins much sooner than in the dynamometer tests. 
In fact, it ends in the former where it begins in the latter. One can 
therefore hardly say that it is a matter of muscular fatigue. The first 
explanation to suggest itself is a lessening in interest, and this is 
strengthened by the fact that the drop occurs at that time when he was 
most affected by the monotony of the routine work. In this test less 
depends for improvement upon the increase in muscular power than 

Whipple's Manual of Mental and Physical Tests, Baltimore, 1910, p. 101. 



THE PSYCHO-PHYSIOLOGY OF A FAST. 



195 



in the dynamometer tests, the main factor being the rapidity of action. 
We know that the rate of the reaction time is greatly affected by changes 
in attention, and it is probable that the betterment in the muscular 
control, which we may assume from the results of the dynamometer 
tests did occur, was insufficient to offset this loss of interest. The 
results of the last days confirm this assumption, for here we undoubt- 
edly have the effect of interest in an end spurt, which, notwithstand- 
ing the muscular fatigue undoubtedly present at this time, brings the 
curve back to a higher level. In regard to the two plateaus referred 
to above, it seems plausible to infer, from what we know of the causes 
of plateaus in the learning process in acts of skill, that these sudden 



A/o.of 



")ays i 



OS 






































































\ 












































































1 






\ 












It 
































































/ 


\ 






-.. 










» 

6 

4- 
1 

e 
-l 






. 


























, 


i 




/ 


\ 










.' 














1 










\ 














\ 






/ 


K 


"" 


^ 












J 


•. 


. 


' 


\ 






. 




-'' 


\ 




s 


1 


















.-■ 












\ 
















i 


^ 


\ 




;' 


\ 


/ 








"-. 


' 






\ 


/ 




\ 
























'■•-, 










i 










\ 


/ 




\ 




/ 




- 


















/ 




1 
































\ 




















s 


' 


























































\ 




















































1 














































































































s 
















































































































































































no 




































































































































































Ho 
m 


























































































































































































































































































IBS 


































































































































































M 

m 




































































































































































































• 

















































fast began 



It It « 13 14- If IS 17 IB 13 iO U II a 2* tS i6 tl li i3 30 SI 3i 33 3* / 

t 

Fast ended L afer fesfs 



Fig. 24. — Tapping tests. 



rises to new levels are due to the learning of some new method or short 
cut. Here the most obvious short cut is the lessening of the height 
of the stroke. 1 

An examination of the difference curve (IV, fig. 24), which has been 
obtained by subtracting the result of the last 10 seconds from that of 
the first 10, still further confirms the assumption of a wavering in in- 
terest. There is a gradual increase in the amount of this difference, 
which indicates fatigue. This increase is particularly marked toward 
the end, when the records are improving, which means that the improve- 
ment is caused by a spurt during the first 10 seconds. 

'This is an error which is bound to occur with this form of tapping-board. The writer has 
therefore, recently constructed a board which regulates the height of the stroke, thus making it 
a constant factor. 



196 



A STUDY OF PROLONGED FASTING. 



In general, it may be said that although initial lack of interest 1 
and later muscular fatigue played a role, both factors being directed 
toward a decrease in the amount of work, yet the will impulse toward 
the end was sufficiently great to bring the curve back to its initial level 
and almost to its maximum. 

STRENGTH TESTS. 

These tests immediately followed the tapping tests. The subject 
stood and received the dynamometer, one of the Collin type, from the 
experimenter, and pressing it, returned it to the experimenter. The 
record was noted and the instrument returned. The interval between 
trials was about a second. Ten trials were made with the left hand, 




Days 113 4 

fasf began 



7 8 9 16 II It 13 I* 15 >t IT 18 19 U tl IZ « Z* 15 U 17 18 19 80 31 it )i 3* I t 3 * 5 6 

fast ended Later tests 
Fig. 25. — Strength tests. 

followed by ten trials with the right. Both in the right-hand (VII, 
fig. 26) and left-hand (V, fig. 25) curves there is an initial falling off, 
which is more marked with the right hand. The left-hand curve, 
however, continues to fall to the tenth day, when it takes a de- 
cided drop, while the right-hand curve declines more gradually to the 
ninth day, when it reaches its minimum. Both curves then rise to a 
maximum, which is reached by the left hand on the sixteenth day and 
by the right hand on the twelfth day (the record of the first day not 
being considered in speaking of this maximum). The curves then 
fall, the left much more than the right, especially in the middle of the 
series, the former reaching its minimum on the thirty-first day. Both 

'Against this suggestion is the fact that other tests did not show this lack of interest, but it is 
quite possible that the interest varied with the different tests. 



THE PSYCHO-PHYSIOLOGY OF A FAST. 



197 



curves show a slight end spurt. This is, as a glance at the curve will 
show, merely a rough picture, there being decided rises and falls 
throughout. 

In interpreting the curve it must be remembered that L.'s left hand 
is the practiced hand and it can therefore be assumed that the muscles 
of that hand are the stronger. In fact, the results make this more than 
an assumption, for the record of this hand is at all times decidedly 
better than that of the right hand. The initial falling off is what one 
must expect when the subject is not accustomed to the particular 
muscular exercise. There is a great exertion at first, and the muscles, 
skin, and subcutaneous tissue feel the unusual strain for several days. 
The muscles least accustomed to exercise are the most affected, and 



Lbs. 
95 








































































l 
















































































\ 


















































































\ 


















































































\ 






























































\ 






i 






L. 




















































/ 


s 


/ 






X 
















/ 


\ 


/ 


\ 


/ 














































V 










_\ 




/ 


V 










/ 


± 


/ 


\ 


/ 


























































^ 


v 


/ 


\ 


f 


\ 






/ 
















































\ 


















i 






























































\ 
















70 
































































\ 
















































































\ 
















20 


























































,' 










i 








,, 






" x 










12 










/ 




\ 


; 










> 














^ 


,-■ 




■ v 


» 


s 


/ 


\ 


i 


\ 


















... 




! i 




/ 




\ 


/ 


\ 


,' 




*«. 


/ 




^„__ 


X 


r . 


/ 


















\ 




























i ; 


.^ 


/ 




' 




1 


t 




■ 






























































pt— 

































































































































































, Oays i z j + 5 6 7 j s in ii it a i* is i6 it i« if to ii it 23 24 is it a u » so it si 35 34 i imm 

fast ended Later tests 



fast JbegfO/? 



Fig. 26.— Strength teats. 



for this reason the right-hand record drops more than that of the left 
hand. Then the muscles gradually recover and the effect of practice 
begins to appear. Acting against the practice is the increasing fatigue. 
The right hand being the unused hand gives practice more chance for 
its influence and although fatigue never allows the curve to reach its 
first day's record, yet the drop which soon begins is much more gradual, 
as has been pointed out, than it is with the left hand, which shows 
more clearly the effect of fatigue. 

The difference curves (VI and VIII), which were obtained by sub- 
tracting the average of the last three records of each day from the 
average of the first three, help to strengthen the conclusions just 
drawn. The rise of the difference curve at the same time as the fall 
of the main curve means, of course, increasing fatigue, which shows 
itself in a greater and greater drop toward the end of the daily series. 



198 



A STUDY OF PROLONGED FASTING. 



This rise in the two difference curves is relatively about the same, 
which means that the daily increase in fatigue is relatively the same 
for the two hands. Further, if we glance at curves IX and X, fig. 
27, we find additional indications in the same direction. This curve 
is plotted from the first of the daily series of 10 trials. This trial is 
least affected by fatigue and therefore shows the greatest influence of 
practice. Here there is a gradual rise for the right hand until next to 
the last day, while the curve for the left hand begins to drop where 
it should according to our analysis. 

In general, we may therefore say that fatigue appears in both hands 
early in the series. The curve for the left hand drops far below the 
record of the first few days. The curve for the right hand shows less 
drop, due to the greater influence of practice, so that the two curves 
tend to approach one another. 



Lbs. 




Days I z 3 * S « 7 t s i» u « u i* is te 17 it 1* to ti ei a « zs i* er it » 30 31 3i 33 3* / 1 3 4- s 6 
Fas? began fas/ ended Later tests 

Fig. 27. — Strength tests. 

TACTUAL-SPACE THRESHOLD. 

A pair of dividers with wooden tips were used as an sesthesiometer. 
The threshold was found on the volar side of the forearm, about 4 
inches from the elbow. The points were applied on either side of a 
red-ink dot which was made on the arm on the first day and renewed 
when necessary. The method of minimal change, with ascending and 
descending series, was employed; 5 trials, excluding one-point "vexier" 
trials, were made at each distance; 4 correct out of 5 was considered 
the threshold. 1 

For the first few days the curve (XI, fig. 28) keeps the high level of 7 
cm. On the seventh day there is a drop to 5.5 cm., then a slight rise to a 
level of 6 cm. and a high threshold of 6.5 cm. on the fourteenth day, 
followed by a fall to the minimum of 5 cm. on the twenty-second day, 
which minimum is again reached on the twenty-sixth and thirtieth 

x It had been intended to call 3 out of 5 the correct threshold, but this was not found feasible. 
The threshold is probably too high, but for the present purpose, where the change and not the 
absolute threshold is being investigated, this does not matter. The curve shows no record for the 
fourth and fifth days. The experimenter was absent on these days and the physician who kindly 
volunteered his services did not deem himself sufficiently skilled in this particular test to undertake it. 



THE PSYCHO-PHYSIOLOGY OF A FAST. 



199 



days. The final days show a rise to 6 cm. The decided drop on the 
seventh day may be due to adaptation to the experiment, which in this 
instance means the adoption of a definite and clear criterion of dis- 
crimination. The drop in the middle of the series, after a more or less 
constant level, may be due to a similar cause — that is, a change to a 
better criterion. The rises in the latter part of the curve are never as 
great as those of the first part, although on the last day the curve 
again reaches 6.2 cm. This threshold had to be placed at 5 correct 
judgments, as there was a jump from 3 correct judgments. This makes 
the threshold probably too high. If we omit the first day and compare 
the average of the period from the seventh to the twentieth day with 
the average of that from the twenty-first to the thirty-fourth day we 
find a difference of 0.4 cm. in favor of the latter period. We may say 




Days I S 3 4 3 6 7 8 4p}/0 II IZ 13 !■*■ IS 16 h IS 13 10 Zl ZZ 23 Z4-ZS ZS Z7 18 Z3 30 31 32 35 34- I £ 3 + 5 i' 

t 
Fasf ended L afer /esfe 



fxzsf began 



Fig. 28. — Tactual-space threshold and visual acuity. 



then, in general, that there is a very slight improvement in the dis- 
criminating process, but that there is no end spurt, which latter, from 
the very nature of the process under investigation, is not to be expected. 



ROTE MEMORY FOR DIGITS. 

The usual rote memory test was employed. Increasing series of 
digits, beginning with 4 digits, were read aloud once by the experi- 
menter to the beat of a metronome with 1-second intervals and were 
repeated as far as possible by the subject. The combinations of digits 
varied daily. 

Curve II B, fig. 23, is obtained by taking the last series that con- 
tains only one mistake, curve II A, fig. 23, by taking the number which 
immediately precedes the one containing the first mistake. Curve A, 
which gives a picture of the rote-memory process, shows two apexes of 
maximal value near the middle and another on the thirty-first day. 



200 A STUDY OF PROLONGED FASTING. 

There is, however, a very low minimum in the second half of the curve 
and a decided drop from the maximum of the thirty-first day. One can, 
therefore, hardly speak of an improvement. The most that can be 
said is that toward the end of the fast the subject was again able to 
reach the maximum record of 10 digits obtained near the middle of 
the series. From curve B we see that on the third day a mistake was 
made at 4 digits, yet the retention is 9 digits; on the eleventh day a 
mistake at 4 digits and a retention of 8, etc. It seems fair to assume 
from these results that curve B represents in a rough manner the degree 
of attention. It is only inattention that can produce results like the 
above. Curve B shows a decided rise to the eighteenth day, when it 
reaches a maximum, and although it follows a lower level from this day 
it never reaches the minimum of the first third of the series. One may 
therefore say that there is an improvement in the state of attention, at 
least for this experiment, as the fast progressed. 

ASSOCIATION TESTS. 

The free-association experiments consisted of the daily presentation 
of a list of 20 words, which were selected principally from the lists pre- 
pared by Woodworth and Wells, 1 and with the exception of the list 
of May 9, which was a repetition of that of April 11, they were all 
different. 2 Several days after the tests were begun it was thought 
advisable, in order to make the lists as uniform as possible, to have 
them composed of an equal number of verbs, concrete nouns, adjectives, 
and abstract nouns, in the order given. This arrangement was 
adhered to from April 18 to the end of the tests, with the exception of 
May 9. The words were read aloud by the experimenter and the time 
taken with an ordinary stop-watch. The reproduction experiments 
followed these with only a pause of a minute. Although the subject 
was told that he need not repeat the same word, if it did not come at 
once, yet there is little doubt that his efforts were always directed 
toward that end. L. had a good command of the English language, 
although it is not his native tongue, but at times he had difficulty in 
finding the word he wanted. In such cases he made a gesture as soon 
as the idea came to him and the watch was snapped at that time rather 
than when the English word was found. This method of procedure 
was not often necessary and it seemed a legitimate means of balancing 
the slight disadvantage he had as a foreigner. A reserve list was pre- 
pared upon which to draw when he did not understand the word of the 
main list. 

The curve (XIII, fig. 29) is plotted from the daily average. The 
average was used in order the better to include the influence of the 

'Woodworth and Wells, Association Tests, Psych. Monog., 1911, 3. 

2 The lists will be found in Appendix n, pp. 222-229. In a few instances the same word appears 
in two lists. 



THE PSYCHO-PHYSIOLOGY OF A FAST. 



201 



long times, which might very well be ol importance in these tests. 1 The 
few exceptionally long times, such as 20 seconds, which may have been 
caused by emotional complexes, were not included. 

The curve begins with very long reaction times. L. had never 
performed such tests before, so that the sudden drop on the third day 



■Sees. 



Oays 




12 3 +56 

r 

fast began 



7 8 5 10 II 12 12 1+ IS 16 17 18 IS 20 21 22 23 24- 25 26 27 28 29 30 51 32 33 3-* 

t 

fast ended 
Fig. 29. — Free association tests. 



2 3 •+■ 5 6" 

Later teste 



Sees. 














































J 


\ 










l.l 








































I 


\ 




/ 


\ 


r JCIV. Assoc/at/on react/on to 










Z.O 




































i\ 




\ 






\ 










1.3 




































I 


\ 


1 


\ 




/ 


\ 






















Jl 










IS 






































\ 


1 




\ 


/ 


















I 


\ 
















1.7 
















1 






















\ 


i 




\ 


j 




d 














/ 


\ 
















16 
























\ 


























i 














/ 












\ 






1.5 
















i 






1 








{ 


\ 


















r 


























\ 






1.4 
12 


















\ 














\ 




I 














T 


/ 












/ 








I 






\ 






























l 






\ 
















\ 


/ 








\ 




/ 










1 




\ 
























j 


\ 




/ 






















J 


/ 


























\ 




LI 

1.0 




















\ 




\ 


/ 




















































\ 




















































































Z.I 
















\ 






1 


| 


























































20 














/ 


\ 






/ 
















/ 


\ 










































1.8 
1.7 
1.6 
AS 
1.4 

a 

12 
l.l 

1.0 






















/ 














i 


/ 


\ 


I 


\ 








/ 


\ 








































/ 


























\ 




\ 




/ 


s 


/ 




\ 










































\ 


1 






1 


I 


j 


\ 








\ 




\ 




/ 












/ 


\ 














































\ 




\ 




1 






j 


\ 


A 












\ 


/ 


\ 


/ 


\ 










































\ 


































\ 


/ 


\ 






































1 


1 


1 


' 
















\ 






























1 




f 




















\ 


/ 


\ 




/ 








































































\ 


/ 




\ 


1 








































































1 


/ 










Days 


i 
fa 


S3 * S 6 7 » 9 10 II IZ 15 14- 15 it 17 18 19 10 11 22 Z3 £•* 25 Z6 27 28 29 30 31 32 33 34- / 2 3 * 

T t 
sr /began fast ended L ater 

Fig. 30. — Association tests. Reactions to verbs and nouns. 


+es 


- 
ts 


5 



J The median, which was also calculated, gave the same general curve. 



202 



A STUDY OF PROLONGED FASTING. 



must be attributed to the practice improvement, which at this early- 
stage could very well be sudden and of considerable amount, rather 
than to the fact that it is the first day of the fast. From this point the 
curve descends with a few breaks to the fifteenth day, when it reaches 
1.4 seconds; it then rises to the twenty-second day, when it reaches the 
maximum (if we do not consider the first few days) and then falls to the 
end of the series. On the second from the last day it reaches the min- 
imum of 1.3 seconds. Also the record of 1.4 seconds is obtained 3 times 



5+cs 




QayS I 2 3 ■*■ 5 6 7 8 9 10 U 12 15 <4 15 K5 17 16 19 20 a/ 22 23 24 25 26 27 26 29 30 31 32333+12 3 4-56 

t 

fa sr ended L ater fesfs 



fast began 



Fig. 31. — Association tests. Reactions to adjectives. 



Sees 
26 

ts 

14 
1.3 

22 
LI 

ZJO 
1.9 
1.8 
f.7 

1.6 

w 

IA 
1.3 

12 
II 



Days I _ 







































































































































































































1 


, 






























































/ 


\ 




























































































































































/ 








\ 
















































/ 








\ 








































































\ 
















































































\ 






Y 




























































\ 
















\ 


/ 










/ 


\ 














\ 




































\ 
















\ 


/ 




\ 






/ 


\ 




/ 


\ 








\ 




































\ 






























\ 


z 




\ 




; 




\ 




\ 














/ 


\ 
















\ 




































\ 




! 




\ 


1 


\ 












/ 




\ 




















J 
































\ 


/ 










\ 


7 


\ 








/ 




\ 


















\ 














































\ 


f 


\ 




s 


/ 






\ 






































\ 




























\ 
















































































































































































1 i 




J ' 


f 


5 


5 


r 


? i 


i 


1 


i i 


Z 1 


3 / 


f 1 


S 1 


e i 


7 I 


J / 


9 e 


oz 


/ 2 


2 £ 


32 


4-2 


5£ 


6 Z 


7 2 


8 2 


9 I 


3 


1 3 


?! 


33 


4- 


i 


I 


' 


y 1 


1 ( 





Fas? began 



Fia. 32. — Association tests. 



fast ended Later fesfs 
Reactions to abstract nouns. 



in the second half of the series. If we include the first few days it can 
be said in general that there is a very decided betterment in the associa- 
tion times; and even if one calculates from the third day there is an 
appreciable drop. Especially interesting is the almost steady improve- 
ment shown in the last third of the curve. 

In order to analyze the curve further, separate curves (XIV, XV, 
XVI, and XVII, figs. 30 to 32) have been plotted for each of the four 



THE PSYCHO-PHYSIOLOGY OF A FAST. 



203 



categories of stimulus words. It must be remembered that these 
curves begin on the seventh day, when this division into separate 
categories was first made. In consideration of the fact that the daily 
average is obtained from only 5 reactions, too much importance must 
not be attached to sudden daily falls and rises, such as in the abstract 
series on the nineteenth and twentieth days and in the adjective series 
on the eighteenth day, etc., but rather the convex shape of the verb 
curve, the rise in the middle of the noun curve, etc., must be considered. 
It is evident that the rise in the main curve about the tenth to thir- 
teenth day is caused largely by the noun curve and that the relatively 
greatest improvement at the end of the curve as compared with the 



Sees. 


/ 


\ 
























































/ 


\ 








' 
















































/ 


\ 
























































/ 


\ 






































A 












































\ 








/ 


\ 




/ 


\ 






/ 


\ 








/ 


\ 
















































\ 


/ 


\ 




/ 


















\ 


7 


\ 


/ 


\ 


/ 


\ 














































































\ 


1 


\ 




/ 


\ 




/ 


\ 






















































































\ 


















































































\ 


















































































\ 




]/ 


































































































(9 


/ 


1 














































































1.8 


/ 


















































































/ 
















































































1.6 


/ 






































































































/ 
































































/ 
































































/ 


\ 












I 


\ 


























/ 


\ 




































/ 




\ 














\ 


I 


\ 






















/ 


\ 




















1 
















/ 




\ 




/ 


v 










I 


\ 




/ 




\ 














/ 










































' 


/ 




\ 


/ 


















\ 






/ 






\ 


/ 


















.J 




















































\ 




/ 










































































































— 



Days I Z 5 4- S 6 7 9 9 10 II IZ /J 14- /S IS 17 16 IS to Zl £z £3 Z-t tS « 17 it 39 30 31 $Z 45 Jf / 2 3 4-5 6 

* T T 

Fast began Fast ended L ater tests 

Fig. 33. — Reproduction tests and mean variations. 

beginning is in the abstract curve. On the other hand, the verb and 
noun curves have several low averages in the beginning that were not 
reached again. In fact, it is hardly possible to say that either of these 
curves shows general improvement; certainly not the noun curve. An 
examination of the daily fluctuations in the curve shows that the 
fluctuation becomes less as the tests progress. 

The curve (XVIII, fig. 33) for the mean variation of the main curve 
shows a decided improvement as the fast progresses, with a very low 
level on the last 3 days. 

The reproduction curve (XIX, fig. 33) follows the tendencies of the 
association curve. There is the initial drop and many more high peaks 
in the first two-thirds of the series. If it were not for the rise on the 
last two days the general betterment would be more marked. The reac- 



204 



A STUDY OF PROLONGED FASTING. 



tions were, on the whole, rapid, averaging about 1 second and dropping 
as low as 0.8 second. As the number of false reproductions was very 
small (see I, table 21), amounting to only 23 in 680 reactions, or 3 per 
cent, and never more than 3 in one list, an improvement or the reverse 
in this respect would mean little. At least one can say that the quality 
of reproduction suffered no deterioration with the progress of the fast, 
but that retention was equally as good at the end as at the beginning. 

Table 21. — Qualitative analysis. 







II. Classification of reaction words in association 


III. Mistakes in 








experiment. 




cancellation test. 


No. of test. 


I. False 
repro- 


























ductions. 


Misun- 


Identity. 


Persev- 


Repeti- 


Word com- 


Omis- 


Incor- 
rectly 
crossed. 






derstood. 


eration. 


tion. 


pounding. 


sions. 


1st 


1 








1 







1 


2d 


2 


1 


1 








1 





Fast began. 


















3d 


3 


1 
















4th 


1 










3 


2 


1 


5th.... 


2 




1 








3 





6th 







1 






1 


2 





7th 











3 




1 





8th 







i 








1 





9th 





i 










2 





10th 


1 




i 








3 


2 


11th 


1 












2 


1 


12th.... 


1 




l 














13th 















2 


1 


14th 











i 




1 





15th 


1 


















16th.... 













2 


1 


1 


17th.... 





l 








1 


1 





18th 















4 





19th 





l 




1 




1 


1 


1 


20th 


2 




2 














21st 


1 




l 






2 








22d 


2 


















23d 


1 








1 




1 





24th.... 





l 










1 


2 


25th 





















26th 


1 




2 






1 


1 


1 


27th 







1 








3 





28th 







1 






1 


1 


1 


29th 







1 














30th 















1 


1 


31st 





















32d 







1 






2 








33d 


2 




2 




1 




4 


1 


Fast ended . 


















34th.... 


1 




2 














Later tests: 


















1st 









. . 


1 


1 


4 





2d 







3 








1 





3d 







1 




1 










4th 









1 










5th 


1 






2 


1 







6 


6th 









1 


1 




1 






THE PSYCHO-PHYSIOLOGY OF A FAST. 



205 



The quality of the association reactions was of high grade throughout 
the main test (II, table 21). There were no senseless or pure sound 
reactions and very few repetitions. Synonyms, word-compoundings, 
and misunderstood stimulus words occurred seldom and were scattered 
throughout the days. The word "woman" appears a number of times 
and "man" slightly less often. There was also evidence of a religious 
complex. An examination of the different categories did not show suffi- 
cient change to warrant an analysis or tabulation as to quality. It was 
thought that the introduction of words designating food might pro- 
duce delayed reactions both with the word itself and the words immedi- 
ately following. This was not the case. For example, on April 16 we 
find egg-white 1.4 seconds; on April 19 omelet-eat, 1.4 seconds; on 
April 21 fish-sea, 1.4 seconds; on May 7 candy-sweet, 0.8 second; on 



Sees. 






















1 1 1 J I 1 1 1 1 1 






\ 
























c2 






















-JC2C. Association reaction noun-verb - 




i 


















1 








2.1 


























\ 


1 






















10 




















































i 
























1.9 








































A 1 








\ 




\ 


i 























IH 






































"7VT7T 










\ 


\ 


















1.7 




















/ 


^ 










1 A 


TT 














\ 


\ ^ 




1 
















16 


















■ 




X 








JlLA 


/ 












\ 




\ 


/ 


\ 




/ 
















1.5 


















1 




3 






/\ 




























/ 






\ 


/ 


\ 












1-4- 


















I 






\ 




/ 




































\ 


/ 




\ 












A3 


























\ 


/ 




































\ 


/ 




^ 


. 




/ 






12 


























V 




































] 








\ 




/ 






U 
















/ 


































































1.0 



















































































Lays I I 3 + S 6 7 9 9 la II II 13 I* IS U 17 IS 19 iO tl 12 Z3 Z* t5 it 17 16 19 30 31 31 33 34- I Z 3 * 5 tf 

Fasf began Fast ended Later tests 

Fig. 34. — Controlled association tests. 

May 9 apple-fruit, 0.8 second; on May 10 roast-meat, 1 second; on 
May 13 chocolate-sweet, 1 second. None of these reactions were 
followed by unusually long reaction times. It might be of interest to 
mention at this point the unusually long reactions which point to com- 
plexes. On April 13 we find pulse-hand, 9 seconds; on April 21 death- 
eternal, 22.4 seconds; and on April 26 uncertainty-pendulum, 12.6 
seconds. These are the only extremely long reaction times. The 
next longest is 6 seconds. All of these delayed reactions may be 
explained from the same cause. L. had asserted that the chief factor 
for a successful fast was faith and confidence and absolute lack of fear. 
He thinks it is the fear combined with exposure which causes death 
in shipwrecks and other calamities where food is not obtainable, and not 
the actual lack of food. It is also claimed that those who fast frequently 
cover their mirrors in order that they may not be disturbed by the 
evidences of emaciation. One of the supposed dangers in fasting is 
heart failure. If L.'s heart had shown alarming symptoms the fast 
would have been terminated at once. It does not, therefore, require 



206 A STUDY OF PROLONGED FASTING. 

a stretch of imagination to suppose that L. would keep his mind from 
such subjects as death and uncertainty and that he would even avoid 
thought of the condition of his heart and that the mention of these 
words would cause hesitation. 

The determined association reaction noun-verb was begun on the 
eighth day. Curve XX (figure 34) resembles that of the verb curve, 
except that the rise continues longer. It starts very low (1 second), 
increases with rather large daily fluctuations, and on the last day of the 
fast returns to 1.1 seconds. A particularly disturbing factor in this 
series was the fact that there was an ever-increasing difficulty to obtain 
appropriate words. At first the words had obvious associations. 
They were names of common objects, such as dog, gun, eye, etc., but 
more unusual words had to be employed in increasing numbers, and 
there seems no doubt that this circumstance was at least partly the 
cause of the increasing length of the reaction time. It is even more 
important in the determined than in the free-association experiments, 
to have the quality of the words the same and not more difficult. 
For long series of tests the free-association experiments are much to 
be preferred. 

CANCELLATION TEST. 

Special forms were made for this test, consisting of type-written pied 
text of 100 a's and 50 of each of the other letters of the alphabet. A 
different combination was made each day, so that the subject should 
not become accustomed to the order. L. was requested to cancel all 
the a's. He used his left hand and the time was taken with a stop- 
watch. Special care was observed to have the illumination constant 
and the same pencil was employed. 

The curve (XXI, fig. 35) represents the time for the completion of the 
task. As in some of the other curves, so here we have the initial rise, 
which continues to the sixth day, when there is a sudden drop to a 
level which slopes slightly to another sudden drop on the twenty-ninth 
day and a very low level for the final days. The difference between the 
maximum of 3 minutes 48 seconds on the sixth day and the minimum 
of 1 minute 53 seconds on the last fast day is very considerable. The 
maximum is over double the minimum, and even if we compare the 
minimum with the initial time of 3 minutes 7 seconds or with 2 min- 
utes 43 seconds of the seventh day, which is the first and largest 
practice drop, we still find a very considerable difference. There does 
not seem any doubt, therefore, that there is very much of a betterment 
in the time as the fast progresses and that this decrease in the time 
continues to the end of the series. Nor is this improvement in time 
gained at the sacrifice of accuracy. At no time were there many 
mistakes made (see III, table 21). In fact, the degree of accuracy was 
always so high that we can not place any importance on the slight 



THE PSYCHO-PHYSIOLOGY OP A FAST. 



207 



increase of accuracy in the last half of the series 1 nor does the slight 
loss of accuracy at the minimum alter the significance of that result. 



T/'/neT 



— 








f 


t 






































































/ 


s 


/ 












































































/ 


"^ 


7 




























































































































































































































































































































































































































1 


\ 




1 








































\ 


































/ 


\ 




\ 








































\ 


f 
































/ 


\ 




\ 
















^ 


\ 
























































1 


\ 






















\ 




























































-~. 




















\ 


















































































.— 




\ 




































































1 












\ 






























































/ 


s 




/ 












\ 




























































\ 


/ 


\ 




/ 








































































^ 


' 




\ 


/ 
































































































1 . 
Fa 


'1 
ft 


> - 

• 
be 


7<?/ 


r < 
? 


t 


7 t 


i 


t 1 


1 1 


1 1 


Z 1 


3 / 


* / 


J- / 


S I 


7 / 


3 1 


i i 


l 


i i 


Z i 


3 2 


* Z 


5 2 


6 11 28 Z9 30 31 32 33 3* 

r 
Fast ended 


Later tests 


f 



3 5 
3* 



2-40- 

est 



tit 



i o 

Czys 



Fio. 35. — Cancellation tests. 

VISUAL ACUITY. 

These tests were made in the large calorimeter room adjoining the 
small room in which the previous tests were conducted. The largest 
E which had been cut from the Schnellen test-card was used. This 
was held by the experimenter at the level of the subject's head when 
seated. It was illuminated by an electric lamp held by a second 
experimenter in front of the card and moved with it. The shades of 
the room were kept drawn during the experiment in order to have 
constant illumination as far as possible. The subject suffered from 
myopia and wore corrective glasses. A distance was first chosen well 
within the threshold at which the subject was asked to judge in what 
one of the four possible positions the E was being held. The experi- 
menter put the card behind his back when he changed its position. 
After a few days of the tests it was thought that the subject might be 
using the secondary criterion of the distance of the edge of the E from 
the edge of the card, the E not being exactly in the middle. The card 
was therefore mounted on a larger cardboard of the same color in order 
to obviate this possibility. On account of the surprising results, both 
experimenters were at all times keenly attentive to the possibility of 
other secondary criteria, but none could be discovered. Ten trials 
were made at each distance, the card being moved from the observer 
in steps of one foot. That distance was considered the threshold 
which preceded the distance at which the subject made two mistakes 

^here were 29 mistakes in the first half and 24 in the second half of the series. 



208 A STUDY OF PROLONGED FASTING. 

out of the ten trials. 1 The alteration in the position of the E followed 
no definite order, but every means was used in this respect to confuse 
the subject in order to remove all possibility of his guessing the position. 
Most of the judgments were made without hesitation, both at the very 
low and very high thresholds. 

The curve (XII, fig. 28) represents the daily threshold in feet. 
There is a very rapid rise from the fifth to the fourteenth day, when 
the maximum of 37 feet is reached. Then there follows a drop to 24 
feet and a rise to 36 feet on the next to the last day of the fast. The 
thirty-fourth day shows a drop to 19 feet. The record of the fifth 
day is 16 feet, which is the minimum; that of the thirty-second day is 
36 feet, which is 1 foot less than the maximum. This difference of 20 
feet is very great for visual acuity. He saw twice as far at the end 
of the fast as he did at the beginning. 

LATER TESTS. 

Owing to an attack of colic resulting from the nature of the first food 
taken after the fast and the subsequent withdrawal from the labora- 
tory, it was impossible to continue the tests during the recuperative 
period, as had been planned. Only by later tests for comparison could 
a decision be reached as to the efficacy of fasting. One year after the 
tests just described L. volunteered as subject for a short series of tests. 
These were conducted at the Harvard Psychological Laboratory and 
extended over a period of 6 consecutive days. It was not possible to 
arrange for them to take place at 5 o'clock as previously and 10 o'clock 
in the morning had to be chosen. All the other conditions were 
observed as closely as possible. The same tests, with the exception of 
the visual-acuity test and the hand-writing test, were performed. 
L. seemed in good health. He weighed about 126 pounds, which is 
somewhat less than he weighed when he began his fast. His physical 
appearance was, however, very much the same as on the day he arrived 
at the Nutrition Laboratory. He had remained in America during 
the previous year, engaged in medical studies, lecturing, etc., had 
not fasted again, and had had no illness during that time. In com- 
ing to the laboratory he made a journey of 4 miles and had already 
had several hours' work, having risen each day at 5 o'clock, exercised 
for half an hour, and made several visits. The conditions previous to 
the tests are, therefore, hardly comparable to those of the former series. 
It is evident, however, that he was as strenuous, if not more so, than he 
had been up to the later hour of 5 o'clock of the previous tests. 

The rote memory for digits (II a) was somewhat poorer than it was 
during the latter part of the long series. It did not reach the maximum 

1 Lack of time prevented the threshold being taken in the reverse direction. The tests took 
5 to 10 minutes. 



THE PSYCHO-PHYSIOLOGY OF A FAST. 209 

by two numbers, yet it did not show any poor scores. The curve (II b) 
which represents the first mistake, or, as it was supposed above, the 
state of attention, shows an improvement over the latter part of the 
first series in that it does not drop so far. On the other hand, the rote 
memory for words (I a) seemed as good if not better than during the 
fast. It reached the former maximum on the fourth day and never 
dropped below eight words. The memory after 55 minutes (I b) was 
as good as the immediate memory. From these results it may be 
concluded that the memory was still, after the year's interval, at about 
the level that it was at the end of the fast. 

The curve (III) for the tapping begins considerably higher than the 
maximum of the fasting tests, and although it drops somewhat, still it 
remains above the former maximum. The drop in the difference 
curve (IV) is caused principally by a falling off in the initial spurt. 
This is concluded from the fact that the results of the last 10 seconds 
vary much less than those of the first 10. 

The results of the first day of the tactual space threshold can not be 
utilized as a comparison (XI). The unusually high threshold was 
undoubtedly caused by inattention on the part of L., who admitted 
that he had been very much worried over an appointment he had been 
forced to miss and upon which his mind had been during these tests. 
Apart from this day the curve has the same form it had during the latter 
part of the previous trials. The second and third days show the mini- 
mum, which was last reached on the thirtieth day of the former trials. 

The dynamometer used in the previous tests could not be obtained 
until the second day. There are, therefore, only 5 records. The 
curves for both the right hand (VII) and the left hand (V) begin with 
very high records and drop considerably on the second day, just as they 
did in the former series. These first records are very much better than 
any made in the previous trials. Even after the drop the right hand 
twice surpasses the previous maximum and remains close to it on the 
other days. The difference curves (VI and VIII) show that on the first 
day the high record for the left hand was made by a sustained effort. 
The right-hand spurt caused fatigue toward the end. The large dif- 
ferences during the next 3 days for the left hand were caused by spurts 
followed by fatigue, that of the right hand by fatigue. It is seen that 
the strength of the muscles of the hand had very much increased since 
the end of the fast, and judging from the first day's results was much 
greater than at the beginning of the fast. One acquires a knack in 
gripping the instrument and it may be that this was carried over from 
the former tests and made these initial records higher than those of the 
year before. In other words, some of the effect of practice was still 
present and influenced the results much more than it did when it had 
the opposing effect of fatigue. 



210 A STUDY OF PROLONGED FASTING. 

The free-association reaction time (XIII) begins at the low point 
of the last day of the previous series; on the third day it reaches the 
shortest time of that series, and again on the fifth day and on the last 
day it falls almost one-fifth of a second below this point. That is, the 
curve continues the descent it began in the middle of the former series 
in as regular a manner as if a year had not intervened. Inasmuch as 
some practice was necessary after so long an intermission, it may be 
said that the reaction times were better than they were at the end of 
the fast. The m. v. (XVIII) was 0.5 second on the first day and 0.15 
second on the sixth, with an almost steady decline. 

The average reproduction time (XIX) is 0.9 second for all the days; 
this is very low, and although 0.8 second was reached 3 times in the 
former tests, it is safe to conclude that the reproduction times are at 
least as good as they were at the end of the fast. In fact, the average 
for these days is better than for any 6 consecutive days of the previous 
tests. There was only one false reproduction and that was "wrong" 
for "bad." In view of the fewness of the trials little would be gained 
by an analysis of the results according to categories XIV, XV, XVI, 
XVII. The noun and adjective curves are lower than the verb and 
abstract curves. The quality of the reactions is about the same. 
Evidence of a religious or mystic complex is as plain here as in the pre- 
vious results. "God" was the reaction for "adore," "worship," 
"unseen," "mercy," "Divine," and "Infinite;" "supreme" gave 
"Being," "sacred" gave "church," "adorable" gave "saint," "life" 
gave "eternal," and "ornament" gave "church." There were no very 
long reaction times. In connection with the previous complex it may 
be mentioned that "death" was the reaction word for "fear." 1 

The reaction noun-verb (XX) begins at the average of the thirty- 
second day of the former series and on the third and fourth day reaches 
the minimum of the next to the last day of the long series. The average 
of these days is very much better than that of the last days of the fast 
series or even of the first days, so that there is no doubt of an improve- 
ment in these reactions. 

The cancellation test (XXI) begins at about the point of the twenty- 
seventh day and the time gradually decreases, but at the sixth day 
has not reached the rapid time of the last fast day; but judging from 
the slope of the curve, one would expect it to do so shortly, so that one 
can conclude that the mental functions necessary for this test are in 
about the same state they were at the end of the fast. There were only 
6 mistakes, 4 of them being on the first day. 

*It was thought that a year's intermission would make the old lists equivalent to new ones and 
as one would then be sure of having the lists of this series of the same quality with those of the 
former, the old lists were used on the first day, but 7 of the 20 reactions were the same as those 
made a year ago, so that new lists were made. 



THE PSYCHO-PHYSIOLOGY OF A FAST. 211 

CORRELATIONS. 

It would be supposed that there would be very good and very poor 
days upon which all the curves would show proportionate increases or 
decreases, or that at least similar tests, such as those of the higher 
mental processes, would show similar variations. If we compare 
some of the crests and valleys, however, we arrive at negative results. 
For instance, on the twenty-second day the association time (XIII) 
is long and both memory curves (I a and II a) are in a valley, but the 
cancellation test (XXI) shows improvement, and the reproduction 
times (XIX) are not long. On the sixteenth day the left hand reaches 
a maximum in the strength tests (V), but the right hand (VII) shows 
no such result. Tapping (III) rises on that day, but it is still compara- 
tively low; one memory curve has fallen (II a) and the association time 
(XIII) has risen. On the fifteenth and seventeenth days the memory 
curve (II a) is at a maximum and the association time (XIII) is also lower; 
the cancellation test (XXI) is also low on these days, but the maximum 
of the memory tests (I a, II a) on the thirty-first day finds the associa- 
tion times (XIII) longer. On the twelfth day the curves for the 
strength tests (V, VII) have risen for both hands — it is the maximum 
for the right hand — the time for the cancellation tests (XXI) has short- 
ened and memory (I a) is better, but the tapping record (III) has fallen 
and both association (XIII) and reproduction times (XIX) are at a 
peak. The considerable lengthening of the time of the cancellation 
test (XXI) on the sixth day finds a betterment in most of the other 
tests, the tapping test (III) indeed, having reached its maximum on 
that day. The visual acuity curve (XII) rises abruptly to its maxi- 
mum on the fourteenth day and, although with a few exceptions the 
curves show a slight betterment, the rise is comparatively insignificant. 
It must be concluded, therefore, that with the exception of the last 
day the daily fluctuations can not be traced to any one cause, such as 
a general bodily fatigue and depressed mood or vigorous and cheerful 
mental states, but that either there is a change in the one or more 
processes essential to the particular test that is showing the exceptional 
rise or fall or that there has been a momentary wave of fatigue or dis- 
traction or spurt, etc. A diary of the fast was kept in which every 
important incident was noted and it is possible that many of the fluctu- 
ations in particular curves or changes in general tendencies of several of 
the curves could be more or less satisfactorily explained. The follow- 
ing considerations, however, make such explanations of doubtful value. 
One can not say in advance what the effect of visits or other changes 
in the general routine may be. Much depends upon the particular 
circumstances. Now, if the results were better after a certain visit, 
one could say that the subject was in a pleasant mood after the break 
in the monotony of the days and that his mind had been stimulated by 
agreeable conversation. If the results were worse on those days, one 



212 A STUDY OF PROLONGED FASTING. 

could say with equal weight that the fatigue following the unusual 
exertion was the cause. Only the most reliable introspection on the 
part of the subject before and after each test could have given strength 
to such explanations, and both the lack of time and training on the part 
of the subject made such a procedure impossible. 

It did seem possible, however, to make an exception of the days on 
which L. took a drive or was allowed on the roof and that if the curves 
showed an agreement in their fluctuations on these days an unequivocal 
explanation could be found. The drives were taken on the fourteenth, 
seventeenth, twentieth, twenty-second, twenty-fourth, twenty-ninth, 
thirty-first, and thirty-second days; the visits to the roof on the tenth, 
fifteenth, twenty-first, and thirtieth days. As was stated above, there 
was no general agreement even on these days. In regard to the individ- 
ual curves, however, the visual acuity curve seemed to show the influence 
of the drives. The best result in the visual acuity test was made on the 
first drive day and the curve always ascends on the drive days, although 
not always to a peak. It falls, however, on all but one day when a visit 
was made to the roof; that it rises on the drive days is contrary to what 
one would expect and is difficult of explanation, since the subject's eyes 
should, if anything, have been fatigued by the increased light. If there 
had been a stimulation of the central processes causing a heightened 
power of discrimination, this ought to have influenced the other curves 
as well. 

GENERAL SUMMARY AND CONCLUSIONS. 

The fact that a human being could live for a month or longer without 
food had already been satisfactorily proved. 1 Merlatti is reported 
to have fasted for 50 days and Dr. Tanner for 40 days. The fast of 
Succi 2 is most similar to that of L. in that it was undergone for about 
the same length of time and under similarly strict scientific control, 
although never before had quite so many precautions been taken as 
in the case of L. Succi fasted for 30 days, but took pepton on the 
twenty-seventh day. L. continued for one day longer, absolutely 
nothing but distilled water passing his lips during that time. Both 
men remained in good physical condition throughout and seemed at 
no time to suffer any unusual discomfort. It was with difficulty that 
L. was persuaded to discontinue his fast on the thirty-first day. 
Although Luciani doubted that Succi was mentally normal, general 
observations and the tests pointed to a sound mind in the case of L. 
Both men were, naturally, men of great determination and above all of 

*E. Bardier, in his article "La Faim" (Ch. Richet's Dictionnaire de Physiologie, 1904, 6, p. 3), 
remarks in regard to voluntary and involuntary fasts: "On pourra se soumettre volontairement 
a un jeune prolonge, comme l'experience en a plusiers fois ete tentee, et endurer assez facilement 
les souffrances de la faim. Le besoin de manger sera d'autant moins douloureux, d'autant plus 
facile a supporter qu'il suffira d'un signe pour etre mis en face d'un succulent repas. Au contraire, 
la faim sera beaucoup plus penible, ses manifestations beaucoup plus douloureuses, si Ton se croit — 
dans un naufrage, dans une expedition — voue a une inanition complete sans espoir de salut." 
On page 6, in reference to forced fasting, he further says: "La lutte que Ton est oblige de soutenir 
contre les causes m6mes de cette inanition augmente la sensation de faim." 

2 Das Hungern, by Luigi Luciani. Translated into German by Dr. M. O. Fraenkel. 1890. 



THE PSYCHO-PHYSIOLOGY OF A FAST. 213 

implicit faith and confidence in their idea. L. believed fasting to be a 
panacea for all ills and the very fact that he is of that type of man who 
can narrow his horizon about an idea and stubbornly resist all inva- 
sions gave him the best equipment for the fight against the natural 
demands of the flesh. Such a type of mind can not be called abnormal, 
although it is unusual. The feeling of hunger was at all times, even 
during the first stages of the fast, denied by L. This statement should 
not be disbelieved, even though frequently there is extreme discomfort, 
which those who fast tell us only disappears after the second or third 
day, as in the case of Succi. With L. and perhaps with other fasters 
this feeling of hunger may have been suppressed from the beginning 
by auto-suggestion. The fact of the deep-ingrained faith in the fast 
makes this plausible. 1 

The condition of Succi's higher mental processes was only ascertained 
by general observation. These observations agree with those made 
upon L. There was at no time any symptom of hallucination or 
lack of clearness in the thought processes. Luciani writes: 

"Am 13 Hungertage wollte ich seine Ausdauer bez. geistiger Anstrengungen 
auf die Probe stellen, indem ich ihm schwierige oder unlosliche metaphysische 
und theosophische Fragen vorlegte und bestandig Einwtirfe gegen seine Ant- 
worten erhob, in der Absicht, seinen Verstand zu ermiiden. Ich muss geste- 
hen, nicht bemerkt zu haben, dass sein Geist dabei mehr ermudete als der 
jedes andern Sterblichen von gleichem Bildungsgrade und gleicher Begabung, 
wenn man ihn solchergestalt martert." 2 

L. is a man of a much higher level of intelligence and intellectual 
training than Succi. At all times during the fast he was very eager to 
enter into discussions upon abstract subjects such as the value of the 
Esperanto language, the political conditions in Malta, the possibility 
of mental telepathy, and theories of spiritism, as well as the value of 
fasting. It could not be observed that there was any diminution of 
his argumentative powers or lack of lucidity of expression. When 
aroused to counter argumentation he showed the same energy in reply 
at the end as at the beginning of the fast. 

Succi's muscular strength as well as his sensory acuity was ascer- 
tained in a manner somewhat similar to the method employed for L., 
and the results will be compared in the following summary and inter- 
pretation of results: 

(1) In the dynamometer tests made upon Succi it is impossible to 
tell from the text how many trials were made daily. As the curves for 

X E. Bardier, in criticizing Bernheim, writes: " Au sens oil l'entend Bernheim, les jeuneurs qui se 
soumettent a l'inanition resistent facilement, tout simplement par le fait d'une auto-suggestion. 
Discutant en particulier le jeune de Cetti, il admet que ce dernier—tout en n'etant pas un hysterique— 
s'est suggestionne. II demeure convaincu qu'il conservait toute sa force physique, 'cela suffit pour 
realiser le ph6nomene; l'idee fait l'acte; il s'exalte, il s'entratne, il se nourrit de son idee, il se 
montre avec complaisance a ses visiteurs, il jouit de son triomphe; l'esprit domine le corps; 
etc' . . . Le jeftneur, par sa volonte, arrive a resister a l'habitude de manger; il obdit a sa 
conscience qui le soumet a l'abstinence, mais certainement sa volont6 doit 6tre incapable de provo- 
quer la suppression d'une sensation." La Faim in Ch. Richet's Dictionnaire de Physiologie, 1904, 
6, p. 10. See also footnote 3, p. 191 of this publication. 

2 Luciani, Das Hungern. German translation by Dr. M. O. Fraenkel, 1890, pp. 68-69. 



214 A STUDY OF PROLONGED FASTING. 

the 10 trials and for the initial trial for L. are similar, the 10-trial curve 
will be considered. It is safe to assume from lack of mention of the 
fact and from the nature of the curves that Succi was right-handed. 
It will therefore be necessary to compare the curve of the right hand of 
Succi with that of the left hand of L. 

It will be remembered that the strength of both hands was found to 
increase after the drop on the second day until the right hand (VII, fig. 26) 
reached its maximum on the twelfth day and the left hand (V, fig. 25) on 
the sixteenth day, both curves then dropping steadily from this point, 
the right, however, less than the left, for the left reached a minimum 
on the thirty-first day, while the right during the fast never dropped as 
low as the record of the nineteenth day. There is a very striking 
similarity between these and Succi's tests. 1 Both of Succi's curves 
also drop after the first trials and then rise again, his left reaching a 
maximum on the fourteenth, his right hand on the twentieth day, as 
compared to the twelfth and sixteenth days of L. Succi's curves then 
drop also, but the left drops more than the right, which is the reverse of 
L.'s curves. With Succi both maximums are greater than the first 
day's records, while with L. this is the case with only the left hand. 
This agrees, however, with L.'s records for the initial daily trials (IX 
and X, fig. 27). Further, L. was able to make a spurt at the end of the 
fast with both hands, this spurt extending through several days. Succi 
was only able to spurt with one hand and that on the last day, the curve 
for the other hand remaining stationary. 

Luciani attributed the rise of the curve alone to auto-suggestion. It 
seems quite probable, inasmuch as Succi and possibly L. also believed 
that their strength would be increased by the fast, that this idea strength- 
ened their determination and that they bettered their results by sheer 
"will power." 2 There is, however, another possibility which maybe 
assumed without denying the influence of auto-suggestion, namely, 
that, at least in the case of L., who was unused to such tests, the coor- 
dination of the muscles became gradually more perfect, and further, that 
these muscles, which were being exercised daily, increased for a time in 
strength as they would have done under normal conditions, but in this 
case possibly to the detriment of other muscle groups. In both cases, 
with both hands, fatigue gained the ascendency over practice effect 
and possibly over auto-suggestion about the middle of the fast, causing 
the curves to drop. In the case of L.'s unpracticed hand, however, the 
effect of practice had more room to work and held the curve up longer 
than in the case of the practiced hand. 

(2) The tapping test (III, fig. 24) is also influenced by the condition 
of the muscular tissue, but there is another factor more essential here 

luciani. Das Hungern, 1890, p. 55. 

2 E. K. Strong, Jr., in his paper entitled " The effect of various types of suggestion upon muscular 
activity" (Psych. Rev., 1910, p. 278), says: "The auto-suggestion tends most strongly of all the 
types of suggestion to heighten the maxima." 



THE PSYCHO-PHYSIOLOGY OF A FAST. 215 

than strength, and that is the reaction time. As in the strength tests, 
there is a rise at first, but here it is of much shorter duration, the maxi- 
mum of 215 taps in 30 seconds being reached on the sixth day. 
The following considerable drop until the fifteenth day, at a time when 
the strength tests are showing more efficiency, may possibly be caused 
by a lessening in the interest for this test. 1 About the middle of the 
series this interest and increased effort for a good record may have 
returned, judging from the results, but fatigue had by that time set in 
and the curve, although rising until the last day, is never quite able to 
reach the maximum of the sixth day; that is, there was some falling off 
in the rapidity of reaction, which, judging from the results of the strength 
test, was due rather to a change in the muscle tissue than to a change in 
the nervous arc. 2 From what we know of the effect of practice in 
such tests it is most probable that if it had not been for this increased 
muscular fatigue the curve would have reached an appreciable maxi- 
mum at the end of the series. From the fact of the very small differ- 
ence between the average of the first 10 and last 10 seconds on the sixth 
day, when the maximum was reached, as compared with the great 
difference in the almost equally good result of the last day, it is evident 
that on the first day the good performance of the first 10 seconds prac- 
tically continues throughout (in both instances the best record was 
made during the first 10 seconds), while on the last day the effect of 
practice as shown in the initial performance was counterbalanced 
toward the end by fatigue. 3 These results seem to cast further doubt 
upon Luciani's hypothesis of auto-suggestion in the strength test, for 
surely auto-suggestion should play as great, if not a greater, role in 
the tapping tests during those days in which according to the strength 
tests it would have to be assumed at work. The results of the tapping 
tests are indeed directly opposed to such a theory. 

To sum up, it may then be said that though initial lack of interest and 
later muscular fatigue played a role, both factors being directed toward 
a decrease in the amount of work, yet central factors toward the end 
brought the curve back to its initial level and almost to its maximum. 

J See pp. 194 and 196. 

2 As the tapping tests preceded the strength tests, the objection can not be raised that the hand 
was being unusually fatigued by these latter tests. 

In reference to the tapping test under normal conditions, Wells writes that "The objective 
fatigue phenomena which we note in the test are in all probability fatigue phenomena in the refrac- 
tory phase or a lowered efficiency of coordination, equally a product of altered synaptic conditions; 
the sensations of fatigue, on the other hand, may with equal assurance be ascribed to tissue changes 
within the muscles that take place as a result of their continued effort." (F. L. Wells. Normal 
performance in the tapping test before and during practice, with special reference to fatigue 
phenomena. Am. Journ. Psych., 1908, p. 473.) In the above tests the change in muscular tissue 
is due to emaciation, a fact that does not play a role in the test to which Wells refers. At no time 
did L. speak of sensations of fatigue, and judging alone from his facial and bodily expressions 
there are no data from which to assume that they were greater at the end than at the beginning 
of the fast. As to the synaptic conditions, there is nothing in the test to point to a change. 

3 Wells writes: "The true practice gain is one mainly in the initial efficiency of performance, as 
distinguished from the warming-up gain, which shows itself chiefly in continued efficiency of 
performance." Am. Journ. Psych., 1908, p. 478. 



216 A STUDY OF PROLONGED FASTING. 

(3) The threshold for tactual-space perception (XI, fig. 28) decreased 
somewhat as the fast progressed. It was on the average much better 
during the last half than the first half of the series. Similar tests were 
made upon Succi upon a number of different parts of the body, but only 
on 3 days, before the fast, on the fifteenth day, and on the twenty-ninth 
day. On some parts of the body there was an increase, on other parts 
a decrease. Luciani believed the difference in the 3 days due to differ- 
ences in degree of attention. On that part of the body corresponding 
most closely to the spot used in these tests, *. e., the lower third of the 
volar side of the forearm, there happened to be a rather large decrease 
in the threshold, the three thresholds being respectively 16, 11, and 
10 mm. 1 Authorities differ as to whether practice lowers the threshold 
in tests performed under normal conditions. Dresslar, 2 for example, 
found that practice had a considerable effect. Solomons 3 found that 
if the subject is not informed of his errors there is no effect of practice. 
In the above tests the subject was never told of his mistakes and 
"vexier" trials were introduced at frequent intervals and in no special 
order, yet there was a lowering of the threshold. This may be and prob- 
ably is due to several causes. A physiological cause would be a decrease 
in the fat, thus exposing the nerve endings and making them more 
sensitive. On the psychological side increased attention, which we 
find indicated in other of the tests, would lower the threshold for dis- 
crimination. Also, as the tests progress the image of the criterion used 
becomes cleared. From what is known of the process of perception, 
this is a most important factor in explaining the above effect of practice. 
The physiological change is the only one which could be attributed 
unequivocally to the fast. The central change occurs in series under 
normal conditions. 

If, as has often been assumed, the tactual space threshold test is a 
measure of mental fatigue, then it must be concluded that there is no 
indication of such fatigue during the fast. 

(4) The visual acuity (XII, fig. 28) showed an astonishing better- 
ment. From 17 feet as the distance of clear vision for the particular 
test card employed, the curve ascended rapidly to 37 feet on the four- 
teenth day and, although there is a falling off, 36 feet is the record for 
the last day of the fast. 

If it were not for the maximum of 37 feet midway in the series, 
the improvement would be comparatively a steady one. One explana- 
tion that suggests itself is that the possible change in intra-ocular 
tension caused the eye-ball to change its shape. Unless his glasses 
were not the proper ones for him, however, a change in the eye should 
cause more rather than less difficulty as long as he wore his glasses. 
Further, the suddenness of the rise seems to vitiate such a theory. 

'Luciani, Das Hungern, 1890, p. 64. 

2 F. B. Dresslar. Studies in the psychology of touch. Am. Journ. Psych., pp. 313-368. 1894. 

3 L. M. Solomons. Discrimination in cutaneous sensations. Psych. Rev., pp. 246-250. 1897. 



THE PSYCHO-PHYSIOLOGY OF A FAST. 217 

A satisfactory explanation seems difficult to find. It might be said 
that the 37-foot record was made by chance. This also seems pre- 
cluded by the fact of the number of previous steps in which 10 correct 
answers were given and from the evidence of confidence displayed 
by the subject. 1 

Succi's eyes were examined with the ophthalmoscope and his acuity 
measured before the fast and on the fifteenth and twenty-eighth days of 
the fast, but no change was detected. 2 If L. had happened to be 
measured on the third, sixteenth, and one of the days toward the end of 
the series only, the change would have been thought as negligible as in 
the case of Succi. In all such tests where the daily fluctuation is consid- 
erable three tests in a month are not sufficient upon which to base a judg- 
ment as to the change in sensory acuity or higher mental processes. 

(5) The rote memory for digits (II, fig. 23) showed very little change. 
There is a slight suggestion of improvement during the first half of 
the series. Judging from the curve which indicates the point at which 
the first mistake was made (II b), one can say that there was a gradual 
improvement in this respect, especially in the first half of the series, 
which is probably in part due to a betterment in the perception of the 
spoken word, but especially to an increase in attention, it becoming 
more sustained as the fast progressed. The rote memory for sense 
words (I a) showed a greater improvement than did that for digits. 
Here probably the practice effect consisted in the forming of associa- 
tions between the words. The most marked improvement of all is 
in the retention after a longer period of time, i. e., after 55 minutes (I b). 
This is probably also due, in part at least, to the more frequent forming 
of associations. Besides, the repetition of the same task through so 
many days undoubtedly strengthened the determining tendency, i. e., 
the determination taken at the time of memorizing for the words to ap- 
pear in consciousness again, it remaining either in consciousness or 
subconsciousness during the interval. According to L.'s statement, 
his mind did not revert to the task within the hour. Indeed, the other 
tests followed each other so rapidly that this would have been a diffi- 
cult thing to do. 

Experiments upon memory under normal conditions also show the 
effect of practice, as evidenced by an appreciable increase in the mem- 
ory span which may continue for a period of 2 months. 3 

x The subject did not know whether he was right or wrong or how many correct answers consti- 
tuted a threshold, so that the results could not have been prearranged by him; and if they could 
have been he would not have allowed such a good record on the fourteenth day. The high thresh- 
old on the last day is obviously due to his unusually poor physical condition (when, if at any time, 
one might be justified in speaking of a lack of effort). 
2 Luciani, Das Hungern, 1890, pp. 66-67. 
3 T. L. Bolton. The growth of memory in school children. Am. Journ. Psych., 1892, pp. 362-380. 

G. Milller and F. Schumann. Experimentelle Beitrage zur Untersuchung des Gedachtniss, 
Zeitschr. f. Psych., 6, 1894, pp. 81-190, 257-339. 

W. H. Winch. The transfer of improvement in memory in school-children. British Journ. 
Psych., 1908, pp. 284-293. 



218 A STUDY OF PROLONGED FASTING. 

(6) The cancellation test (XXI), which employs to a greater degree 
the higher functions of perception and attention shows the greatest 
improvement of any of the tests used. This improvement continues 
from the sixth to the last day of the fast. The accuracy is so high 
throughout the series that the slight improvement in the latter part 
of the tests is of no significance. Experiments have shown that fatigue 
affects the accuracy, so that again we have evidence against an increase 
in mental fatigue. 1 

Besides an improvement in the above-named functions, the increase 
in visual acuity may have been a factor in the results. On the other 
hand, from the results of the tapping test and strength tests one must 
conclude that the betterment is in no degree due either to a betterment 
in reaction time or motor ability. 

(7) The free association time (XIII, fig. 29) is on the whole shorter 
during the latter part of the series. If it were not for a rapid drop in the 
middle of the curve after a rise similar to that in the tapping test the 
improvement would be comparatively steady. The minimum of 1.3 
seconds is reached on the day before the last day of the fast and should 
be compared rather with the 1.9 seconds of the third day than with the 
2.5 seconds of the first day, when L. was unaccustomed to the manner of 
reaction. Even when this comparison is made it is seen that the 
improvement is considerable. A separation of the curve into four 
curves corresponding to the four categories used made a more minute 
analysis possible. The curves XIV, XV, XVI, and XVII, figs. 30 to 32, 
show fewer high averages in the second half of the series, but it is only 
in the abstract curve and in less degree in the adjective series that there 
are more low averages in the second half of the curve. In fact, in neither 
of the other two curves is the lowest average of the first half of the series 
again equaled. This seems to indicate that the betterment in the 
general average of the 20 words is principally due to a betterment in 
the reaction to abstract words. It is to be expected that the most 
difficult associations would show the greatest practice effect. In the 
noun and verb curve there is an almost steady rise in the middle of the 
curve corresponding to the rise in the middle of the main curve. I 
seems plausible to suppose that there is here, as in the tapping test, a 
falling off of interest, and that this would manifest itself more readily in 
the easier tasks, in which the reaction is likely to become more nearly 
mechanical. 

The general improvement is also seen in the decrease in the variations 
of the reaction times. In all four curves the daily variation is much 
less in the second half of the series. Parallel with this is the decrease in 

X B. Bourdon. Observations comparatives sur la reconnaissance, la discrimination et l'associa- 
tion. Rev. Phil., 1895, pp. 153-185. A. Binet. Attention et adaptation, Annee Psych., 1900, 
6, pp. 248-404. C. Bitter. Ermudungsmessungen, Zeitschr. f. Psych., 1900, pp. 401-444. 



THE PSYCHO-PHYSIOLOGY OF A FAST. 219 

the variations within each day, as is shown by the decided drop in 
the m. v. curve (XVIII, fig. 33) - 1 

Although the improvement in the reproduction time is not so great 
as in the association time, yet it is noticeable, the average of the second 
half being lower than that of the first, although the very low time of 
0.8 second was made on the second day as well as during the second 
half of the series. 

The quality of the associations was good throughout (II, table 21) 
and showed no striking change. 2 The reproductions were so nearly 
perfect from the first that nothing can be said in regard to them to 
support the results of the memory tests. One might add, however, 
that neither do they contradict those results. 

The controlled reaction noun-verb (XX, fig. 34) shows an increasing 
lengthening of the time until almost the end of the series. It is quite 
probable that this was caused by an increasing difficulty in the stimulus 
words selected, a factor which could not well be avoided. No other 
reason suggests itself why these reactions should have taken a different 
course from that of the free association tests. 

The present methods of testing mental capacity unfortunately do 
not permit one to make dogmatic statements as to the results of any 
such tests. In each one a number of functions are involved, any one 
of which may have produced the variations which occur. For example, 
the cancellation test involves, among other things, attention and inter- 
est, apperception and discrimination, nervous impulse and motor dis- 
charge. But when, as here, a set of tests are employed in which the 
same functions are more or less active and they all show a similar trend, 
then a conjecture along general lines seems legitimate. Further, when 
there is a very decided difference and it is known that a certain function 
is of prime importance, then one is undoubtedly justified in ascribing 
the outcome of this test to changes in this function. It is desired to 
make it plain that no exact measurement is claimed, but merely that it 
has been possible, by means of a number of selected tests, to sketch an 
outline picture of the condition of L.'s psycho-physiological organism. 

1 Wells conducted long series of association reactions with normal subjects and for all of them 
found an improvement in the reaction time. (See Practice effect in free association, Am. Journ. 
Psych., 1911, 22, pp. 1-13.) 

2 W. Weygandt's results are hardly comparable to those obtained in these tests (Ueber die 
Beinfiussung geistiger Leistungen durch Hunger, Psych. Arbeiten, 4, pp. 45-173). His subjects 
fasted for periods of only 24 and 48 hours at a time. This intermittent fasting may possibly 
cause a much more pronounced disturbance to the organism than a prolonged fast. That there 
was greater exhaustion seems to be indicated by the fact that there was an increase in associations 
by sound. He also finds that there was an increase in the outer as compared with the inner asso- 
ciations. (It is now admitted that such a classification of reaction words can not be made without 
introspection.) Weygandt also found memory to be affected. The association time was not 
altered. Aschaffenburg studied the effect on association reactions of the exhaustion produced by 
a night's work without food or sleep. (Studien ueber Associationen, ii Teil. Die Associationen 
in der Erschopfung. Psych. Arbeiten, 2, pp. 1-83) . He too found a similar decrease in the quality 
of the reaction words. "Mit der Zunahme der Erschopfung wirkt die zugerufene Vorstellung 
immer weniger durch ihre Inhalt; an dessen Stelle bestimmen der Klang und die Tonfarbe die 
Reaction." 



220 A STUDY OF PROLONGED FASTING. 

It will be remembered that the tests range from those involving 
principally the muscle groups to those depending in a higher degree 
upon central factors. The test depending most on the muscular 
reactions, t. e., the strength test, showed a falling off. The tapping 
test, which also involved the muscles but in which the rapidity of reac- 
tion was a more important factor, showed no improvement. As soon 
as one turns, however, to the sensory discriminations one notices an 
increased efficiency, which is probably due either to a change in the 
peripheral organs or central processes, or both. Finally all the tests 
involving the higher processes of attention, perception, and association 
show improvement. In a word, there was a loss in muscular strength 
due probably to loss of tissue, a possible gain in sensory acuity and a 
decided increase in the efficiency of all the central processes. It would 
be premature to say that the improvement is the direct result of the 
prolonged abstinence from food, as similar improvement has been 
observed in such tests under normal conditions, due entirely to the effect 
of practice. It can be stated, however, with some degree of certainty, 
that the complete abstinence from food for 31 days had little effect 
upon the higher mental functions, which were able to develop through 
practice very much as they would have done under normal conditions. 

This agrees with the observations upon the physiological conditions. 
It has been found that during a fast the muscle tissues are the first to 
suffer and the nervous tissues the last. From these results it seems 
that up to the thirty-first day the nervous tissues have not suffered. 

These results also confirm in part the general observations made by 
those fasting. It is frequently stated by them that they can do better 
mental work. The results show that at least they can do approxi- 
mately as well, and it is not at all unlikely that some can do better, for it 
must be remembered that there is none of that sluggishness of the 
mental processes directly after eating, when the digestive processes are 
at their height, and there is also absence of indigestion and the after- 
effects of alcohol and tobacco. That, on the other hand, as has been 
often claimed, they are able to do more muscular work and that their 
power of endurance is greater is in this case at least not true. Probably 
the contrast of their actual results compared with what they expected 
would happen to a man without food makes the result seem greater 
than it is. The claim that the senses are more acute has been verified 
as to the visual acuity. It is hardly likely that the slight difference in 
the tactual-space threshold would have been noticed by the faster. 1 

The question remains as to whether prolonged fasting is beneficial 
or dangerous to the organism. This can only be satisfactorily answered 

*L. stated that the heightened sensitivity for odors made walking on the streets of Malta during 
his first fast positively unpleasant. The other senses were examined in the case of Succi and no 
appreciable change discovered. Luciani, Das Hungern, 1890. 

Whipple (Manual of Mental and Physical Tests, Baltimore, 1910, p. 215), in speaking of the 
-effect of practice in the sesthesiometer test, remarks that Dresslar states: "This practice effect ia 
. . . rapidly lost, being reduced very definitely within 8 days and completely lost within a month." 



THE PSYCHO-PHYSIOLOGY OF A FAST. 221 

after an exhaustive physiological examination extending over a long 
period of time subsequent to the fast. The tests made after the lapse 
of a year permit, however, of some conjecture in this regard concerning 
those functions at least which have been discussed in this paper. 

The strength test shows a great improvement over the former record. 
L. exerted a pressure considerably greater than at any time during 
the long series. The record for the tapping test is also above the maxi- 
mum of the previous record. The association test shows a marked 
improvement and the reproduction is also better, especially in that it 
varies less, and the retention of sense words has perhaps also slightly 
improved. The tactual-space threshold and the rote memory for sense 
words are about the same as at the end of the fast. Only in the case of 
the memory for digits and in the cancellation test has the previous 
maximum not been reached, but both of these results show consistently 
good results. It may be stated, in short, that after an entire year's 
intermission the curves continued practically from the point they had 
previously reached, if not considerably above that point, without show- 
ing that loss of practice which might well have been expected. These 
improved conditions are, however, not necessarily traceable directly to 
the beneficial effects of the fast. In regard to the association tests L. 
has undoubtedly become still better acquainted with the English lan- 
guage, and in respect to the strength tests it must be noted that L. has 
exercised his muscles daily, according to his report. In general he has 
led a careful life, paying especial attention to his diet. The possible 
effect of climate and his new surroundings is also to be considered. 
Finally, and most important, is the possibility that there was actually 
a greater effect of practice in the first series than appeared in the 
records, but that it was concealed by certain opposing effects of the 
fast, so that the results of the later tests may not be quite what might 
be supposed from a comparison of the records. 

It remains, however, an indisputable fact that, according to the tests 
made, there was no lasting evil effect of the fast, either upon muscular 
strength or mental activity. 



222 



A STUDY OF PROLONGED FASTING. 



APPENDIX I.-DREAMS. 

As has been already stated, L. was asked to recount the dreams he had had 
during the previous night. From these records those dreams are here given 
which pertain to food. It will be seen that at one time he ate, at another 
refused food, but in neither case was there evi dence of anything but a normal 
emotional reaction. According to the Freudian theory this absence of an 
intense emotional state (there were no nightmares nor anything else in the 
records indicative either of mental or bodily distress) means that the will 
("wish") to fast was too strong to allow of any serious conflict of ideas. A 
great part of the dreams are of a sexual nature and are not here given. 

April 13. I saw a basket covered with a white piece of cloth, which I 
imagined full of food. When I tried to uncover it several black rats jumped 
out of it and frightened me. 

I dreamed I was passing down one of our streets in Malta with a paper bag 
under my arm containing cheese-cakes for my daughter. I found myself in 
a state of mental excitement and after going a certain distance I found that the 
lower end of the bag was opened and the cheese cakes were gone. In their 
stead was a white hand. 

April 19. I dreamed I was in a shop and on the counter there was a very 
big ham, about 10 feet in diameter. The proprietor was riding on the top of it 
with a knife in one hand. "It is a very good one," he said. I answered, "I 
do not like it. Do you not know I am fasting?" Then a friar came in and said, 
"I will take it in his stead, because I like it." He took it and swallowed it. 

April 21. I dreamed I had been for a walk in the country. I went to a 
country tavern and asked for something to eat. The proprietor gave me a 
beefsteak and some fried red fish. I ate them with relish and asked what I 
had to pay. He told me $1.50 and asked if that was too much. I said I did 
not think so. In coming out of the tavern I saw a river full of these red fish 
and people were trying to catch them. I said, " You are fishing out all the 
fish and if you continue you will not have any more to eat." 



APPENDIX II-COMPLETE SERIES OF ASSOCIATION TESTS. 



Average 



2.3 



April 11, 1912: 






April 12: 






Stimulus 


Reaction 


Reaction 


Stimulus 


Reaction 


Reaction 


word. 


word. 


time. 


word. 


word. 


time. 


Paper 


ink 


2.2" 


Round 


table 


2.0" 


Bright 


light 


2.0 


Country 


green 


1.8 


Yellow 


lemon 


1.8 


Silver 


spoon 


2.3 


Table 


knife 


1.2 


Rabbit 


white 


2.0 


Spoon 


broth 


2.8 


Chair 


cushion 


3.0 


Apple 


stem 


2.4 


Glass 


window 


2.0 


Sleep 


bed 


1.6 


Flower 


odor 


2.3 


Room 


door 


1.3 


Sun 


brightness 


3.2 


Face 


eye 


2.0 


Bread 


white 


2.3 


Carpet 


red 


1.8 


Wood 


hard 


3.0 


Animal 


white 


2.6 


Well 


water 


2.4 


Rain 


noise 


5.0 


Danger 


sea 


2.0 


Teach 


bench 


2.0 


Tired 


bed 


2.0 


Doctor 


knife 


4.0 


Watch 


gold 


2.4 


Book 


no. of pages 


3.8 


Marble 


table 


1.6 


Store 


glass window 


1.6 


Iron 


bar 


3.8 


Horse 


tail 


2.2 


Bridge 


iron 


2.8 


Island 


trees 


2.2 


Blind 


dark 


2.4 


Journey 


ship 


2.2 


Pencil 


wood 


3.0 


Freedom 


banner 


2.0 


Candy 


sweet 


3.4 


Sweet 


sugar 


1.2 










Average 



2.5 



THE PSYCHO-PHYSIOLOGY OF A FAST. 



223 



April IS: 






April 16: 






Stimulus 


Reaction 


Reaction 


Stimulus 


Reaction 


Reaction 


word. 


word. 


time. 


word. 


word. 


time. 


Timid 


rabbit 


3.0" 


Defend 


country 


1.8" 


Pulse 


hand 


9.0 


Deck 


ship 


1.2 


Mystery 


religion 


5.2 


Fresh 


air 


0.8 


Savage 


wolf 


2.4 


Faculty 


arts 


1.0 


Spirit 


angel 


2.4 


Deduct 


sum 


1.4 


Teeth 


to eat 


2.6 


Dinner 


good 


1.4 


Bargain 


profit 


5.0 


Flavor 


odor 


2.2 


Blunder 


mistake 


3.0 


Displease 


anyone 


3.2 


Temper 


nervous 


2.2 


Dog 


large 


2.0 


Abrupt 


cascade 


2.0 


Good 


man 


0.6 


Harp 


sing 


2.0 


Fault 


his fault 


3.0 


Switch 


machine 


2.4 


Egg 


white 


1.4 


Wide 


sea 


2.2 


Green 


tree 


1.8 


Tailor 


stuff 


3.0 


Fright 


dog 


2.2 


Income 


money 


1.5 


Drive 


horse 


1.2 


Splendor 


sun 


1.8 


Fairy 


tale 


1.4 


(Salve) solve 


lip 


2.6 


Hard 


stone 


0.8 


Moon 


silver 


2.2 


Function 


ceremony 


3.2 


Frost 


white 


1.8 


Profess 


religion 


1.4 


License 


wine 


1.8 


Salt 


sea 


1.4 




Average 


2.9 




Average 


1.7 


April 14-' 




= 


April 17: 




= 


Accept 


a reward 


2.8" 


Crawl 


serpent 


2.0" 


Air 


blue 


2.6 


Clown 


buffoon 


2.4 


Able 


sailor 


2.0 


Dizzy 


headache 


1.6 


Abuse 


drink 


2.6 


Distance 


my country 


2.0 


Address 


letter 


1.8 


Cure 


physic 


2.6 


Blood 


red 


1.1 


Corn 


grass 


1.8 


Bad 


man --— 


1.4 


Easy 


chair 


1.8 


Age 


90 


1.2 


Distress 


sorrow 


2.0 


Agree 


wife 


1.0 


Decorate 


church 


1.6 


Boot 


black 


1.8 


Copper 


metal 


1.4 


(Tall) baU 


tree 


1.6 


Even 


ground 


2.4 


Balance 


weight 


1.6 


Endurance 


fasting 


1.4 


Amuse 


theater 


1.4 


Decline 


age 


1.0 


Bottle 


ink 


1.4 


Cream 


sweet 


2.0 


Band 


brass 


1.4 


Firm 


strong 


3.4 


Climate 


mild 


0.8 


East 


west 


1.0 


Bite 


dog 


1.5 


Degrade 


man 


1.8 


Box 


wooden 


1.6 


Corset 


woman 


1.0 


Contents 


book 


6.4 


Flat 


floor 


1.8 


Boy 


small 


1.8 


End 


book 


3.0 




Average 


1.9 




Average 


1.9 


April 15: 




= 


April 18: 




= 


Catch 


bird 


1.6" 


Hit 


hammer 


3.0" 


Brain 


human 


2.6 


Swallow 


food 


1.2 


Broad 


street 


1.4 


Suffer 


pain 


1.2 


Courage 


man 


2.2 


Build 


house 


1.3 


Cease 


speak 


2.2 


Rubber 


teeth 


1.4 


Brick 


red 


1.6 


Food 


good 


1.0 


Broken 


glass 


1.0 


Park 


large 


1.1 


Culture 


physical culture 


1.6 


Boat 


swim 


1.8 


Compel 


servant 


3.4 


Smooth 


floor 


1.1 


Cable 


iron wire 


1.4 


Straight 


way 


1.8 


Central 


station 


1.6 


Ugly 


man 


1.8 


Crowd 


people 


1.2 


Gentle 


woman 


1.4 


Confess 


priest 


1.0 


Naughty 


man 


2.0 


Carbon 


carbon dioxide 


1.8 


Power 


England 


1.6 


Common 


sense 


1.0 


Strength 


athlete 


1.9 


Day 


night 


2.0 


Charm 


woman 


' 3.0 


Control 


engine 


1.0 


Cost 


money 


1.0 


Chain 


iron 


1.0 


Kindness 


woman 


2.2 


Course 


study 


2.2 


Break 


glass 


1.2 


Delegate 


apostolic 


2.0 


Jaw 


mouth 


1.8 




Average 


1.7 




Average 


*.6 



224 



A STUDY OF PROLONGED FASTING. 



April 19: 






April 22: 






Stimulus 


Reaction 


Reaction 


Stimulus 


Reaction 


Reaction 


word. 


word. 


time. 


word. 


word. 


time. 


Produce 


field 


1.4' 


Eat 


bread 


2.0' 


Cry 


baby 


1.0 


Open 


door 


1.0 


Freeze 


cold 


1.6 


Divide 


reign 


1.8 


Follow 


soldier 


5.8 


Fade 


flower 


1.6 


Smoke 


pipe 


0.8 


Travel 


ship 


2.0 


Rope 


long 


2.0 


Umbrella 


rain 


0.8 


Omelet 


eat 


1.4 


Gift 


gold 


3.0 


Cap 


head 


1.0 


Man 


long 


0.8 


Burglar 


thief 


1.6 


Sailor 


ship 


1.2 


Delicate 


woman 


0.8 


School 


teacher 


1.2 


Thick 


paper 


2.8 


Dense 


air 


2.0 


Expensive 


money 


1.0 


Short 


man 


1.4 


Dark 


night 


1.0 


Weary 


travel 


1.6 


Unfair 


unjust 


2.0 


Best 


book 


5.4 


Purpose 


Bcope 


1.0 


Excuse 


pardon 


1.6 


Glory 


eternal 


1.2 


Insult 


bad 


3.8 


Mischief 


bad 


2.0 


Prudence 


woman 


1.6 


Occasion 


accident 


1.0 


Caution 


wise man 


2.2 


Nuisance 


wrong 


1.6 


Conceit 


ambition 


2.2 


Overcoat 


dress 


1.0 


Captain 


ship 


1.4 




Average 


1.6 




Average 


1.9 


April 20: 




= 


April 23: 




= 


Prefer 


office 


2.4' 


Collapse 


sick 


2.4' 


Crush 


crowd 


2.0 


Excite 


nervous 


1.6 


Allow 


pension 


1.6 


Begin 


book 


1.8 


Drink 


water 


1.2 


Prosper 


progress 


2.4 


(Solution) 


salt 


2.2 


Hat 


head 


1.2 


salute 






Sister 


brother 


1.0 


Hip 


thigh 


1.2 


Ham 


meat 


2.0 


Lightning 


thunder 


2.0 


Crime 


justice 


2.8 


Parlor 


bedroom 


2.4 


Tight 


shoe 


2.0 


Snake 


serpent 


1.0 


Solid 


stone 


1.8 


Wicked 


man 


1.2 


Cold 


winter 


1.6 


Rich 


millionaire 


1.8 


Clear 


sky 


1.4 


Clean 


body 


1.2 


Hope 


fortune 


3.6 


Bashful 


woman 


1.0 


Dismay 


fear 


1.6 


True 


religion 


5.2 


Offense 


insult 


1.4 


Exchange 


money 


1.0 


Blunder 


mistake 


1.0 


Style 


literature 


1.0 


Future 


time 


4.0 


Power 


gun 


1.0 


Insist 


persist 


2.4 


Result 


good 


1.4 


Trap 


wolf 


2.0 


Nonsense 


foolish 


1.6 


Oblong 


square 


1.4 


Seed 


plant 


1.0 




Average 


2.0 




Average 


1.7 


April 24: 




= 


April 21: 




= 


Restore 


furniture 


1.4' 


Pinch 


pin 


1.4' 


Impress 


printing 


1.8 


Satisfy 


appetite 


0.8 


Flirt 


woman 


1.0 


Nourish 


food 


1.2 


Ask 


question 


1.2 


Drift 


wind 


0.8 


Receive 


letter 


0.8 


Abuse 


drink 


1.2 


Baker 


bread 


1.0 


Ditch 


deep 


1.2 


Athlete 


strength 


1.0 


Tiger 


fierce 


1.0 


Cradle 


baby 


1.0 


Music 


sweet 


1.0 


Bundle 


hay 


1.0 


Fish 


sea 


1.4 


Elephant 


trunk 


1.0 


Death 


eternal 


(22.4) 


Cheap 


money 


3.0 


Soft 


paste 


2.4 


Black 


dog 


0.8 


Ugly 


man 


1.2 


Tender 


meat 


1.4 


Watchful 


policeman 


2.6 


Prompt 


answer 


1.4 


Indecent 


conduct 


3.0 


Ignorant 


man 


1.0 


Haste 


hurry 


1.0 


Confidence 


familiarity 


2.0 


Comfort 


good 


2.0 


Jealousy 


woman 


0.8 


Adventure 


strange 


1.2 


Honesty 


good 


4.2 


Practice 


long 


1.8 


Unbelief 


atheist 


2.4 


Untrue 


falsehood 


1.6 


Heroism 


warrior 


2.0 


Merit 


high 


2.8 










Average 



1.6 



Average 



1.5 



THE PSYCHO-PHYSIOLOGY OF A FAST. 



225 



April 25: 






April 28: 






Stimulus 


Reaction 


Reaction 


Stimulus 


Reaction 


Reaction 


word. 


word. 


time. 


word. 


word. 


time. 


Join 


chain 


1.8" 


Persuade 


argument 


2.4" 


Clasp 


hand 


1.0 


Dig 


ditch 


1.0 


Advance 


pretension 


2.0 


Get 


money 


1.0 


Argue 


discussion 


2.0 


Sting 


bee 


2.2 


Mountain 


large 


1.0 


Preach 


priest 


1.0 


House 


beautiful 


1.4 


Spice 


pepper 


0.8 


Neck 


strong 


1.0 


Star 


Venus 


1.4 


Lamb 


quiet 


1.2 


Ice 


cold 


1.0 


Hero 


brave 


1.2 


Picture 


beautiful 


1.8 


Jealous 


woman 


1.4 


Lip 


red 


1.4 


White 


snow 


2.0 


Easy 


chair 


1.0 


Serious 


man 


1.0 


Unclean 


dirty 


1.4 


Vacant 


space 


1.0 


Red 


rose 


1.0 


Fertile 


land 


1.0 


Rotten 


mud 


2.0 


Reason 


mind 


1.6 


Hard 


flint 


1.0 


Protection 


government 


1.8 


Proposition 


geometry 


1.6 


Solemnity 


festivity 


1.0 


Improvement 


progress 


1.0 


Impudence 


woman 


3.8 


Infamy 


calumny 


2.2 


Convenience 


etiquette 


3.0 


(Competition) 


commerce 


2.4 


Scratch 


nail 


1.6 


competence 











Attraction 


actress 


2.0 




Average 


1 6 









April 26: 




= 




Average 


1.6 


Forget 


memory 


1.2* 


April 29: 




== 


Dislike 


people 


1.0 


Announce 


news 


1.2" 


Prepare 


lesson 


1.0 


Stain 


ink 


1.0 


Admire 


virtue 


1.8 


Finish 


lesson 


1.4 


Protect 


children 


1.6 


Drag 


horse 


2.0 


Starch 


white 


1.2 


Plead 


case 


2.0 


Mutton 


meat 


14 


Cork 


bottle 


2.0 


Ostrich 


feather 


1.0 


Toy 


child 


1.2 


Roof 


house 


2.0 


Key 


door 


1.2 


Little 


boy 


1.0 


Ox 


horns 


2.2 


Funny 


buffoon 


2.2 


River 


water 


1.6 


Gay 


sun 


1.2 


Rusty 


iron 


1.6 


Dead 


black 


1.2 


Ungracious 


bear 


2.0 


Slow 


worm 


1.6 


Irksome 


science 


2.4 


Solemnity 


feast 


1.6 


Equal 


balance 


4.0 


Annoyance 


fly 


1.0 


Late 


hour 


1.2 


Constancy 


virtue 


3.2 


Accusation 


importation 


2.0 


Attention 


mind 


1.4 


Corruption 


money 


2.0 


Uncertainty 


pendulum 


(12.6) 


Poverty 


distress 


3.2 









Imposition 


tax 


1.0 




Average 


1.4 


Adoration 


saint 


1.4 


April 27: 




===== 









Accuse 


judge 


1.8" 




Average 


1.8 


Appear 


star 


2.0 


April 80: 




== 


Polish 


wood 


1.2 


Adore 


saint 


2.2" 


Repeat 


lesson 


1.0 


Perish 


ship 


2.2 


Condemn 


delinquent 


2.4 


Propose 


marriage 


1.4 


Car 


motor 


1.8 


Uphold 


politics 


2.8 


Knee 


leg 


1.8 


Descend 


stairs 


1.2 


Cloud 


white 


1.2 


Slave 


misery 


2.8 


Fun 


joy 


1.8 


Violin 


music 


2.0 


Violent 


wind 


1.2 


(Path) pot 


country 


2.4 


Sour 


acid 


1.0 


Chapel 


church 


1.4 


Dim 


sound 


1.0 


Trumpet 


sound 


1.2 


Condition 


good 


1.0 


Supreme 


being 


1.2 


Deceit 


deceive 


3.0 


Elegant 


woman 


1.6 


Fraud 


wrong 


3.0 


Impudent 


woman 


2.0 


Brutality 


animal 


2.0 


Blame 


offense 


2.4 


Cup 


wine 


1.2 


Gain 


money 


1.0 


Equality 


fraternity 


3.0 


Idea 


noble 


1.0 


Greasy 


pole 


1.2 


Worship 


God 


1.0 


Violet 


odor 


1.0 


Elevation 


spirit 


1.4 









Noisy 


metronome 


2.0 




Average 


1.7 


Level 


ground 


1.0 



Average 



1.7 



226 



A STUDY OF PROLONGED FASTING. 



May 1: 






May 4- 






Stimulus 


Reaction 


Reaction 


Stimulus 


Reaction 


Reaction 


word. 


word. 


time. 


word. 


word. 


time. 


Escape 


prison 


2.0* 


Fast 


long 


1.4* 


Admit 


argument 


2.0 


Dream 


sleep 


2.6 


Joke 


play 


3.0 


Taste 


food 


1.2 


Improve 


mind 


1.6 


Cook 


food 


1.4 


Defy 


enemy 


1.2 


Mark 


ink 


1.0 


Lamp 


fire 


2.0 


Sparrow 


bird 


1.0 


Cabbage 


green 


1.0 


Foot 


large 


1.6 


Paste 


soft 


1.2 


Spider 


insect 


3.2 


Poem 


beautiful 


1.0 


Forest 


trees 


1.0 


Spear 


piercing 


2.6 


Stone 


heavy 


1.0 


Harsh 


sound 


1.2 


Purple 


color 


1.0 


Unripe 


fruit 


1.0 


Infamous 


calumny 


1.2 


Unwell 


sick 


1.0 


Refined 


art 


1.2 


Vile 


fellow 


1.0 


Ungracious 


bear 


1.8 


Admission 


employment 


3.0 


Center 


circle 


1.6 


Thankfulness 


gratitude 


2.0 


Awkward 


gait 


1.8 


Dishonor 


bad 


3.6 


Supremacy 


authority 


2.0 


Intimacy 


friendship 


1.0 


Constancy 


perseverance 


1.6 


Revenge 


fault 


3.4 


Time 


quick 


1.2 


Least 


thing 


2.6 


Gin 


bad 


1.0 




Average 


1.9 




Average 


1.5 


May 2: 




===== 


May 6: 




== 


Deny 


favor 


2.0' 


Invite 


guest 


1.6* 


Burn 


fire 


1.6 


Pin 


clothes 


1.4 


Paint 


wall 


1.8 


(Crumble) 


bread 


1.4 


Betray 


faith 


1.2 


tremble 






Dress 


clothes 


1.4 


Attack 


enemy 


1.2 


Mouse 


black 


2.0 


Wood 


hard 


2.0 


Barn 


corn 


3.0 


Dirt 


nasty 


3.2 


Song 


beautiful 


1.4 


Shoe 


tight 


1.2 


Spider 


feet 


2.6 


Camp 


large 


1.8 


Scarlet 


fever 


1.6 


Cannon 


big 


2.6 


Beautiful 


woman 


1.4 


Ashamed 


fault 


1.4 


Yellow 


fever 


1.8 


Unsafe 


war 


1.6 


Modest 


girl 


2.0 


Raw 


fruit 


2.0 


Wealthy 


man 


2.0 


Smooth 


ground 


1.2 


Justice 


right 


1.4 


Fortune 


money 


1.4 


Trouble 


bad 


2.0 


Disdain 


angry 


2.0 


Quantity 


large 


1.6 


Refinement 


art 


1.8 


Reproach 


fault 


1.2 


Activity 


work 


1.2 


Energy 


force 


2.0 


Accident 


misfortune 


1.6 


Crack 


nuts 


1.0 


Scoff 


offender 


2.4 









Noisy 


clock 


2.0 




Average 


1.8 









May 8: 




===== 




Average 


1.8 


Guide 


a traveler 


6.4" 


May 6: 




== 


Care 


a boy 


2.3 


Dishonor 


sin 


2.4* 


Denounce 


principles 


3.8 


Remove 


furniture 


1.6 


Drop 


stone 


1.4 


Injure 


sword 


2.4 


Suspect 


fault 


2.2 


Plunge 


water 


1.0 


Saddle 


horse 


1.6 


Murder 


thief 


1.4 


Sleep 


bed 


2.2 


Garden 


flower 


1.0 


Fog 


fruit 


1.0 


Nut 


crack 


2.0 


Skin 


animal 


1.4 


Stem 


heraldry 


2.0 


Earth 


ground 


3.2 


Crab 


animal 


2.0 


Rough 


weather 


1.2 


Pickle 


burning 


2.0 


High 


mountain 


1.2 


Noble 


gentleman 


1.6 


Idle 


servant 


1.4 


Nice 


fellow 


1.2 


Humble 


man 


2.0 


Secure 


keys 


1.0 


Active 


boy 


2.4 


Blue 


sky 


2.0 


Health 


good 


1.4 


Swift 


sparrow 


1.4 


Aim 


noble 


1.8 


Disgrace 


fault 


2.0 


Fame 


vain 


2.8 


Security 


policeman 


2.2 


Shame 


wrong 


2.0 


Unhappiness 


marriage 


2.8 


Ability 


great 


1.2 


Rhyme 


poetry 


1.0 









Disaster 


Titanic 


1.2 



Average 



2.1 



Average 



1.7 



THE PSYCHO-PHYSIOLOGY OF A PAST. 



227 



May 7: 






May 10: 






Stimulus 


Reaction 


Reaction 


Stimulus 


Reaction 


Reaction 


word. 


word. 


time. 


word. 


word. 


time. 


Wash 


clothes 


1.0" 


Roast 


meat 


1.0" 


Elevate 


thought 


1.4 


View 


panorama 


1.8 


Deceive 


wrong 


2.6 


Whistle 


a whistle 


1.4 


Ramble 


about 


1.6 


Alarm 


people 


2.6 


Decay 


reign 


1.8 


Indulge 


drinker 


1.4 


Bible 


holy 


1.4 


Frost 


white 


1.4 


Pencil 


lead 


1.0 


Cask 


wine 


1.0 


Crown 


king 


1.0 


Curtain 


silk 


1.4 


Goat 


milk 


1.2 


Nurse 


baby 


1.2 


Candy 


sweet 


0.8 


Ivy 


wall 


1.4 


Restless 


not quiet 


2.0 


Thankful 


grateful 


1.0 


Simple 


countryman 


1.6 


Steep 


stairs 


1.2 


Reckless 


man 


1.2 


Unwholesome 


air 


1.0 


Eternal 


life 


1.2 


Gentle 


woman 


1.4 


Prosperity 


fortune 


1.0 


Faithful 


servant 


1.0 


Jealousy 


woman 


1.2 


Conflict 


nations 


1.2 


Concealment 


to hide 


2.4 


Anger 


bad 


2.2 


Advancement 


progress 


0.8 


Idleness 


vice 


2.4 


Rancid 


butter 


1.4 


Betrayal 


traitor 


1.8 


Honesty 


good 


1.0 


Denouncement 


fault 


2.0 




Average 


1.4 




Average 


1.5 


May 8: 




= 


May 11: 




= 


Deserve 


merit 


1.2" 


Plunge 


water 


1.0" 


Wish 


fortune 


2.4 


Guess 


enigma 


2.4 


Boast 


glory 


3.2 


Rescue 


wrecked 


1.8 


Establish 


manufactory 


1.1 


Believe 


God 


1.4 


Barber 


razor 


1.6 


Carve 


wood 


1.0 


Pebble 


stone 


1.4 


Door 


house 


1.8 


Heart 


beat 


1.2 


Barley 


corn 


1.0 


Machine 


work 


1.4 


Eagle 


bird 


1.0 


Statue 


marble 


1.2 


Chin 


face 


1.6 


Certain 


thing 


2.0 


Pulse 


beating 


1.0 


Natural 


regime 


1.8 


Alive 


man 


1.2 


Correct 


grammar 


2.0 


Exquisite 


sweet 


1.6 


Dusty 


street 


0.8 


Empty 


barrel 


1.2 


Enormous 


building 


1.6 


Bitter 


quassia 


1.8 


Commandment 


God 


1.0 


Lazy 


fellow 


0.8 


Excitement 


nervous 


0.8 


Modesty 


virtue 


1.0 


Restoration 


food 


1.6 


Immensity 


God 


1.6 


Density 


mercury 


1.8 


Preservation 


alcohol 


1.8 


Infirmity 


sickness 


1.8 


Prudence 


virtue 


1.2 


Return 


voyage 


1.6 


Indiscretion 


vice 


1.2 




Average 


1.6 




Average 


1.4 


May 9: 




= 


May 12: 




= 


Paper 


write 


1.2" 


Find 


treasure 


2.0" 


Bright 


sun 


0.6 


Praise 


merit 


2.0 


Yellow 


fever 


1.4 


Pump 


water 


1.0 


Table 


mahogany 


3.2 


Try 


lesson 


1.8 


Spoon 


food 


1.4 


Guard 


tower 


1.8 


Apple 


fruit 


0.8 


Iron 


metal 


1.8 


Sleep 


night 


2.4 


Stomach 


empty 


1.8 


Cut 


animal 


1.8 


Salmon 


fish 


1.0 


Face 


beautiful 


1.2 


Bath 


water 


1.2 


Carpet 


ground 


1.4 


Splinter 


wood 


1.2 


Animal 


fierce 


1.6 


Unfit 


unable 


2.0 


Rain 


weather 


1.8 


Ardent 


fire 


1.2 


Teach 


lesson 


1.8 


North 


south 


2.2 


Doctor 


medicine 


1.0 


Handsome 


lady 


1.2 


Book 


interesting 


1.4 


Price 


high 


2.2 


Store 


goods 


3.0 


Appetite 


good 


1.2 


Horse 


animal 


1.6 


Fable 


^Esop 


2.0 


Island 


Malta 


1.2 


Definition 


grammar 


1.8 


Journey 


long 


1.0 


Queer 


sound 


2.2 


Freedom 


liberty 


0.8 


Ingenuity 


simplicity 


1.4 



Average 



1.5 



Average 



1.7 



228 



A STUDY OF PROLONGED FASTING. 



May 18: 






June 2, 1918: 






Stimulus 


Reaction 


Reaction 


Stimulus 


Reaction 


Reaction 


word. 


word. 


time. 


word. 


word. 


time. 


Distrust 


enemy 


1.8" 


Adore 


God 


1.0" 


Run 


a long way 


2.0 


Perish 


ship 


1.0 


Agree 


friend 


1.2 


Propose 


marriage 


1.1 


Needle 


thread 


1.2 


Uphold 


opinion 


1.8 


Chocolate 


sweet 


1.0 


Descend 


mountain 


1.2 


Twig 


tree 


1.2 


Slave 


poor 


2.2 


Napkin 


white 


1.2 


Violin 


song 


1.6 


Hill 


steep 


1.4 


Brook 


river 


1.0 


Finger 


hand 


1.0 


Chapel 


church 


1.0 


Pretty 


girl 


1.2 


Trumpet 


sound 


1.0 


Contented 


happy 


1.0 


Supreme 


Being 


1.2 


Absent 


minded 


1.8 


Elegant 


lady 


1.2 


Magical 


lantern 


1.4 


Impudent 


boy 


2.8 


Profane 


words 


1.2 


Blame 


fault 


2.1 


Introduction 


to a friend 


1.4 


Gain 


money 


0.8 


Amusement 


theater 


1.2 


Idea 


beautiful 


1.4 


Remorse 


sin 


0.8 


Worship 


God 


1.6 


Calmness 


quietness 


1.2 


Comfort 


pleasure 


3.0 


Nod 


head 


1.0 


Noisy 


room 


1.0 


Calculate 


numbers 


1.0 


Level 


ground 


1.0 




Average 


1.3 




Average 


1.4 


May 14: 




= 


June 8: 




= 


Shock 


electricity 


1.4" 


Cover 


hat 


1.3" 


Sweat 


heat 


1.8 


Hasten 


pace 


1.0 


Melt 


snow 


1.4 


Curse 


son 


3.6 


Stun 


hit 


1.4 


Hurt 


wound 


1.4 


Hunt 


deer 


2.0 


Blush 


young lady 


2.2 


Maiden 


woman 


1.8 


Island 


Malta 


0.8 


Bag 


sand 


2.0 


Copper 


metal 


1.0 


Belt 


leather 


1.2 


Water 


flowing 


1.0 


Cake 


sweet 


1.2 


Lettuce 


vegetable 


1.4 


Unhappy 


miserable 


1.6 


Brandy 


alcohol 


1.0 


Pure 


blood 


1.8 


Unseen 


God 


1.0 


Disorderly 


irregularity 


1.6 


Merry 


happy 


1.6 


Unemployed 


poor 


2.0 


Sacred 


church 


1.4 


Wretched 


miserable 


2.0 


Excellent 


exam 


1.6 


Indulgence 


vice 


1.6 


Adorable 


saint 


1.4 


Agreement 


friendship 


1.2 


Life 


Eternal 


1.2 


Advantage 


benefit 


1.2 


Opposition 


enemy 


1.2 


Injury 


blow 


1.2 


Intellect 


mind 


1.2 


Outrage 


war 


1.6 


Sorrow 


grief 


1.4 


Rubber 


teeth 


1.6 


Education 


school 


1.2 




Average 


1.6 




Average 


1.4 


May 15: 




= 


June 4-' 




= 


Sin 


bad 


1.4" 


Caress 


baby 


1.4" 


Applaud 


merit 


0.8 


Reduce 


salary 


1.0 


Astonish 


marvel 


1.6 


Reward 


behavior 


1.8 


Rejoice 


good news 


2.0 


Talk 


English 


1.0 


Use 


tools 


1.2 


Touch 


table 


1.0 


Spool 


loom 


1.4 


Street 


long 


1.0 


Sheep 


fur 


1.6 


Cane 


reed 


1.2 


Emerald 


precious stone 


1.8 


Soap 


soft 


1.4 


Wagon 


coal 


1.6 


Cheese 


English 


2.0 


Cottage 


college 


1.6 


Drum 


sound 


1.0 


Naughty 


boy 


1.2 


Happy 


healthy 


2.2 


Exacting 


demand 


2.6 


Small 


boy 


1.0 


Thirsty 


man 


1.2 


Difficult 


lesson 


1.2 


Playful 


boy 


1.2 


Painful 


wound 


1.2 


Impulsive 


dashing 


1.8 


Grief 


sorrow 


1.0 


Faithfulness 


dog 


1.0 


Thought 


good 


1.4 


Provocation 


insult 


1.4 


Credit 


great 


1.6 


Contentment 


happiness 


1.0 


Fear 


death 


1.4 


Religion 


faith 


1.0 


Mercy 


God 


1.2 


Profanity 


bad word 


1.0 


Sinful 


man 


1.0 



Average 



1.4 



Average 



1.3 



THE PSYCHO-PHYSIOLOGY OF A FAST. 



229 



June 5: 






June 6 — Continued 






Stimulus 


Reaction 


Reaction 


Stimulus 


Reaction 


Reaction 


word. 


word. 


time. 


word. 


word. 


time. 


Oppose 


enemy 


1.2" 


Infinite 


God 


1.0" 


Enter 


house 


1.2 


Brave 


soldier 


1.4 


Drive 


horse 


1.0 


Ornamental 


church 


1.0 


Lecture 


public 


2.2 


Dreadful 


fight 


1.4 


Flag 


wave 


1.0 


Chance 


good 


1.4 


Ivory 


white 


1.0 


Quarrel 


men 


2.0 


Bed 


sleep 


1.2 


Conscience 


good 


1.2 


Fountain 


water 


1.0 


Scandal 


bad 


1.8 


Pie 


lemon 


1.6 


Evil 


bad 


1.6 


Awake 


morning 


1.4 









Dull 


night 


1.4 




Average 


1.3 


Many 


friends 


1.8 






= 


Green 


leaves 


1.2 


June 7: 






Divine 


God 


1.0 


Irritate 


nerves 


1.0" 


Terror 


enemy 


1.2 


Tame 


animal 


1.0 


Spite 


hatred 


1.4 


Feed 


animal 


1.2 


Advice 


council 


2.0 


Imagine 


vision 


1.0 


Contempt 


enemy 


1.8 


Suffer 


pain 


1.0 


Dispute 


question 


1.2 


Dinner 


good 


1.2 


Telephone 


friend 


2.6 


Raft 


sea 


1.2 









Chart 


fever 


1.8 




Average 


1.4 


Glove 


hand 


1.0 






= 


Bird 


sing 


1.2 


June 6: 






Afraid 


lion 


1.0 


Scold 


child 


1.0" 


Blue 


sky 


0.8 


Walk 


street 


1.0 


Anxious 


desirous 


1.2 


Punish 


criminal 


2.2 


Long 


street 


1.0 


Smell 


odor 


1.2 


Audacious 


hero 


1.2 


Send 


letter 


1.4 


Expression 


vocal 


1.2 


MU1 


flour 


1.0 


Mistake 


great 


1.2 


Elbow 


hand 


1.2 


Devotion 


church 


1.2 


Milk 


white 


1.0 


Errand 


boy 


1.0 


Scissors 


cut 


1.2 


Expense 


great 


1.4 


Moon 


night 


1.2 









Quiet 


night 


1.4 




Average 


1.14 



FECES. 

In the days preceding the fasting period, there was more or less regu- 
lar defecation, but the special interest in the feces in connection with 
this fasting experiment has to do with the defecation immediately pre- 
ceding the fast and that on the days following the fast. After the 
evening meal on April 13, L. had a large defecation, as was noted in 
the history for that day. There was no defecation, however, through- 
out the entire fasting period, as no feces were passed from the time of 
the defecation on April 13 until 5 h 30 m p. m. on May 15, i. e., about 
8 hours after the first food had been taken. It was suggested to the 
subject that it would be desirable, especially on the first day or two, to 
empty the lower bowel with a warm-water enema, but he preferred not 
to do this. 

The defecation on May 15 was coincidental with a severe attack of 
colic occasioned by the taking of an excessive amount of acid fruits, 
which flooded the stomach and the intestinal tract. The defecation, 
which was copious, contained a few hard, well-formed lumps of feces 
about 1 cm. in diameter and with a total length of 6.5 cm. The rest 
of the material was spongy and soft, running like liquid when turned 
from the vessel. The feces had a nauseating odor, necessitating 
frequent access to the outdoor air in transferring and handling the 
material. Another defecation took place about 8 p. m., a third shortly 
afterward, and still another during the night. The feces were all of a 
very soft and liquid consistency, and of a light yellowish-brown color. 
As the hard material was obviously entirely different in nature from the 
soft material it was removed, and probably this alone can here be con- 
sidered as in any way approximating fasting feces. The second and 
third defecations were tested with litmus paper and found to be strongly 
acid, probably due in part to the organic acid present in the fruit juices. 

The fact that there were no feces throughout the 31 days of this pro- 
longed fast is of special significance, as it is commonly stated that fast- 
ing men excrete from 2 to 5 grams of dry fecal material each day. 
In the earlier experiments at Wesleyan University, no evidence was 
found of what might be called strictly fasting feces. In the prolonged 
fasting experiment with L., since the last defecation prior to the fast 
took place only a half hour after the last meal on April 13, at least a 
portion of the feces of May 15 might be expected to result from the food 
on April 13, so that we find it difficult to determine what proportion, 
if any, of the material defecated should be ascribed to the fasting period. 
Unfortunately the exigencies of the situation, especially in view of the 
illness of the subject, made it impracticable to preserve and prepare 
these feces for a microscopical examination. This is much to be 
regretted, as some light might have been thrown upon their source. 
The amount was, however, extremely small, as the total air-dried 

230 



FECES. 231 

material from the hard portion of the first defecation amounted to but 
20.8 grams. 

The amount of fasting feces reported in observations made by other 
investigators is somewhat difficult to explain, except by the fact that 
they are based in large part upon Mueller's observations on Cetti, 
who fasted 10 days. It should be noted, however, that during the 
entire fast Cetti smoked cigarettes more or less and unquestionably 
shreds of tobacco found their way into the alimentary tract. While 
these shreds of tobacco by no means formed the bulk of the fecal 
material, they doubtless stimulated peristalsis, which caused a some- 
what rapid movement along the alimentary tract of the epithelial 
debris and residue of the digestive juices. In view of this probable 
stimulation in the experiment with Cetti it appears fortunate that our 
subject L. did not use tobacco in any form during his fast. 

It is by no means clear whether the weights recorded by Luciani 
for the feces in Succi's fast are for dry material — as interpreted by 
Mueller 1 — or whether the material recovered from the enemata was 
dried down to the consistency of normal feces. Luciani's expression, 
"un residuo solido di consistenza pastosa," 2 would seem to imply that 
the material was by no means anhydrous. With this interpretation, 
the amount of dry material found in Succi's 30-day fast would be not 
far from 37.5 grams or a little over 1 gram per day, instead of 5 grams 
per day, as computed by Mueller. 

In the experiment with L. the fecal material which obviously be- 
longed either to the fasting period or to the food period prior to the 
fast was separated, dried, and analyzed. The results of the analysis are 
as follows: 

Nitrogen 5 . 26 p. ct. 

Fat 21.11 

Fat saponified 25 .42 

Calcium oxide 2 . 859 

Magnesium oxide 1 . 026 

Total weight of dried material 20.8 grams. 

An inspection of these results shows no noticeable difference from the 
composition of ordinary feces, so that we have no chemical indications 
of feces which can be specifically ascribed to the fasting period. Con- 
sequently the only definite conclusion that can be drawn is that during 
the 31-day fast there was no positive evidence of the existence of fast- 
ing feces. 

In this connection, the following report of Dr. Arthur I. Kendall, of 
the Harvard Medical School (now professor of bacteriology in the 
Northwestern University Medical School), on the flora of the intestinal 
tract of our fasting subject, is of special interest. 

Mueller, Archiv f. path. Anat. u. Physiol., 1893, 131, Supp., p. 107. 
2 Luciani, Fisiologia del digiuno; studi sull' uomo, Florence, 1889, p. 37. 



OBSERVATIONS UPON THE BACTERIAL INTESTINAL FLORA OF A 

STARVING MAN. 

By Arthur I. Kendall. 

The question of the composition of the normal bacterial flora of 
adult man has never been satisfactorily settled, although the consen- 
sus of opinion appears to be that B. coli is the form most commonly 
found. The observations recorded below, while not conclusive, 
furnish information which tends to show that at least three organisms 
may persist in the intestinal tract for a month after all food is withheld, 
and in this sense these bacteria are noteworthy. The history of the 
case needs no comment here, other than to state that the subject had 
no food for 30 days prior to the taking of the sample herein reported. 

The material for study was obtained from an enema of sterile physi- 
ological salt solution, 300 c.c. in all, which was injected into the rectum, 
retained for approximately 5 minutes, and recovered in almost full 
volume. The return fluid (collected in a sterile container, with appro- 
priate precautions) was turbid, with but little odor, practically color- 
less, and, except for a very small amount of cell detritus, free from 
particulate matter. No fecal material was recovered. 

The fluid was plated in plain agar, in a dilution of ttttwid while a 
portion (undiluted, and diluted -n/xnr) was examined for anaerobes 
and certain other bacteria. The latter tests were negative. 

The total count on agar plates (in duplicate) was 131 and 133 col- 
onies, respectively, giving a total of 1,310,000 and 1,330,000 bacteria 
per cubic centimeter of washings. Of the 131 colonies, 4 were identical 
culturally with B. mesentericus, 17 were found to be Micrococcus ovalis 
of Escherich, and the remaining 110 were found to be B. coli. B. coli 
was also recovered from fermentation tubes inoculated with 1 c.c. of a 
t: ooo,ooo dilution of the washings, thus confirming the count by the 
plate method for this organism. 

These results, while not striking, are interesting for two reasons : 

(1) Certain bacteria appear to be able to live upon the intestinal 
secretions, even when all food is withheld for at least a month. 

(2) It appears to be impossible to sterilize the intestinal tract by 
simple starvation. This latter consideration should be of clinical 
interest, since it is customary in certain diseases to try to "starve out" 
bacteria from the intestinal tract. 

232 



EXCRETION THROUGH THE SKIN. 

So great is the total excretion from the body, in the respiration, urine, 
and feces, that aside from the sensible perspiration, the skin as a path of 
excretion is rarely considered in any discussion of the loss of body 
material. But leaving the sensible perspiration entirely out of con- 
sideration, the skin plays an important part, for there is cutaneous 
respiration, including both the absorption of oxygen and the excretion 
of carbon dioxide; there is a very considerable insensible perspiration, 
which in its strictest meaning refers to the vaporization of water from 
the skin surface; and there are the excretions of both nitrogenous 
material and chlorides through the skin. 

Although the excretion of gaseous and solid material through the 
skin of the fasting man would normally be expected to be at a mini- 
mum, it seemed desirable, in order to establish sharp balances of the 
nitrogen and particularly of the salts, to determine carefully the cuta- 
neous excretion of soluble nitrogenous materials as well as the sodium- 
chloride excretion. It was not possible to measure the cutaneous 
respiration of our subject in any of the forms of respiration apparatus 
used in the fasting experiment, for in the calorimeter the cutaneous 
respiration is measured with the pulmonary respiration, and with the 
respiration apparatus no provision is made for the measurement of 
the cutaneous respiration. 

The excretion of the nitrogenous material and chlorides through the 
skin as the fast progressed was, however, of particular significance, and 
arrangements were made for determining these. By nitrogenous 
material is meant not the dead cuticle, but the excretion of water- 
soluble material, chiefly in the form of urea. In order to determine 
this accurately, the body of the subject was given a thorough washing 
before the fast. He was then sponged with distilled water and a 
freshly extracted and dried cotton union suit was placed upon him. 
At the end of the week the union suit was removed, the subject was 
again sponged with distilled water, and another freshly extracted and 
dried cotton union suit was given him. The union suit which had been 
removed was then carefully extracted with distilled water and the 
extract water evaporated after the addition of acid. The water in 
which the subject had been bathed was also saved and evaporated 
after the addition of acid. The entire operation was in the skilled 
hands of Mr. T. M. Carpenter. By this procedure it was expected 
that the perspiration accumulating during the week would be absorbed 
by the cotton union suit and the soluble solids, including salts, urea, 
or other material, would be extracted by the distilled water. 

The nitrogen was determined by the Kjeldahl method. The chlorine 
was determined by titration with silver nitrate and sulphocyanate. 

233 



234 



A STUDY OF PROLONGED FASTING. 



The total amount of nitrogen and chlorine found each week is given 
in table 22, in which it is seen that the nitrogen ranged from 0.73 gram 
in the first week to 0.30 gram in the last week, and the chlorine from 
0.39 gram and 0.41 gram for the first two weeks to 0.18 gram in the last 
week. 

It will be noted that as much as 0.1 gram per day of nitrogen 
in water-soluble material may be excreted through the skin during the 
first week of fasting and that in all probability this method determines 
the minimum rather than the maximum amount, since unquestionably 
there is a continual transformation of urea to ammonium carbonate 
with a loss of ammonia. On the other hand, it is probably true that 
the secretory activity of the skin decreased somewhat as the fast pro- 
gressed, as is evidenced by the values for both nitrogen and chlorine. 
This loss of nitrogen through the skin has special significance in con- 
nection with so-called "nitrogen-balance experiments." 



Table 22. — Cutaneous excretion of nitrogen and chlorine in experiment with L. 



Date. 


Nitrogen. 


Chlorine. 


1912. 

Apr. 13-Apr. 20 1 

Apr. 20-Apr. 27 

Apr. 27-May 4 

May 4-May 11 


gm. 

0.73 
.39 
.31 
.30 


gm. 

0.39 
.41 
.23 

.18 



*The subject was bathed on the evening of April 
13 and at the end of each week thereafter. 

It has previously been shown 1 that during severe muscular work as 
much as 200 milligrams of nitrogen may be excreted through the 
skin per hour. If, therefore, the excretion of nitrogen in a fasting exper- 
iment with minimum activity amounts to 0.1 gram or more per day, 
it is obvious that nitrogen-balance experiments which do not take into 
account this loss through the skin will not give accurate results. I 
am unaware of any determinations of this kind made on a fasting man, 
although Zuntz and his co-workers on Monte Rosa recorded the loss of 
nitrogen and chlorine through the skin in their experiments on the high 
Alps. 2 

The amount of chlorine excreted through the skin of L. was relatively 
small, being approximately from 50 to 60 milligrams per day in the 
first 2 weeks of the fast. During the fourth week of the fast only 

Benedict, Journ. Biol. Chem., 1906, 1, p. 263. 

2 Schwenkenbecher and Spitta (Arch. f. exp. Path. u. Pharm., 1907, 56, p. 284) found about 
0.3 gram each of nitrogen and sodium chloride per 24 hours with a healthy person in bed. Taylor 
(Journ. Biol. Chem., 1911, 9, p. 21) found with two men at work but no visible perspiration per 
day 0.028 gram sulphur, 0.003 gram phosphorus, and 0.190 gram nitrogen in one case and the 
corresponding figures for the other were 0.015, 0.002, and 0.160. 



EXCRETION THROUGH THE SKIN. 235 

about 25 milligrams per day were thus excreted. While the loss of 
nitrogenous material from the surface of the skin by decomposition 
might be considerable throughout the week, it is hardly probable that 
any large amount of chlorine would be mechanically lost. Thus these 
values probably represent very nearly the actual cutaneous excretion 
of chlorine during this period. In this connection it is of interest to 
note the recent work of Wahlgren, 1 indicating that the skin is one of 
the principal reservoirs for chlorine in the body. Finally, attention 
should be called to the discussion of the excretion of water-vapor 
through the skin, 2 in which the evidence points towards a decreased 
secretory activity of the skin as the fast continued. 

Wahlgren, Archiv f. exp. Path. u. Pharm., 1909, 61, p. 97. 
2 See page 373. 



URINE. 

Urine analysis has in the past decade undergone a striking revolution 
as a result of the development of unique and exceedingly accurate 
methods by Folin. Formerly clinical examinations of urine included 
urea determinations, usually by the hypobromite method, and quali- 
tative or roughly quantitative estimations of phosphates, chlorides, 
etc., but to-day the intelligent clinician deals only with the 24-hour 
excretion of the various urinary constituents. The introduction of 
the Kjeldahl method did much to advance our knowledge of the con- 
stituents of the urine by giving us information as to the total organic 
nitrogen, but it remained for Folin to show us the methods for the 
partition of the nitrogen in the urine and its significance. The ammo- 
nia, urea, uric acid, creatinine, and creatine in the urine then began to 
be of much greater significance than was the total nitrogen; but in all 
these advances in the development of urine analysis, and particularly 
in the interpretation of the results, we find stress invariably laid upon 
the nitrogenous constituents. To such a degree is this true that we 
are inclined for the most part to think of the urine solely as a path for 
nitrogen excretion. 

Our previous experience with fasting subjects, however, has shown us 
that in the urine we have not only indices of the protein katabolism, 
but that with acetone, diacetic acid, and /3-oxybutyric acid present, 
we have indices regarding the defective fat katabolism; furthermore, 
the inorganic constituents, such as chlorine, phosphorus, sulphur, and 
the alkaline bases, give us evidence as to the mineral metabolism, the 
sulphur excretion also having an importance in interpreting the protein 
katabolism. It was therefore essential to study the urine of our fasting 
subject not only from the standpoint of protein katabolism, but like- 
wise from every other possible standpoint, so that complete analyses 
were necessary. In carrying out such a study of the fasting urine, we 
have depended more largely upon the results of our former study of 
fasting subjects 1 than in any other part of the research. 2 

GENERAL ROUTINE OF COLLECTION AND SAMPLING. 

In order to give us as much information as possible about the pre- 
vious dietetic habits of this man, particularly for the few weeks prior to 

Benedict, Carnegie Inst. Wash. Pub. 77, 1907, pp. 345-419. 

2 After the pages of this book were in page proof, my attention was called to the article from 
Aoyama's clinic in Tokio, by Watanabe and Sassa, entitled "Die Harnanalyse wahrend des zwei- 
wochigen Hungerns eines Mannes" (Zeitschr. f. Biol., 1914, 64, p. 373), issued from Munich en 
August 27, 1914, but not received here until late in November. It is thus impossible to make any 
comments upon this interesting paper. The authors studied body-weight, measurements of the 
body, body-temperature, pulse, respiration, and the blood, but laid special emphasis upon exten- 
sive urine analyses. Their findings are, for the most part, in full conformity with those recorded 
here. 

236 



URINE. 237 

the fasting experiment, L. was requested to measure and sample the 
urine each day from the first of April until he reached the Nutrition 
Laboratory on April 10, preserving the samples with chloroform. This 
he did most carefully, his training as a pharmacist assisting him materi- 
ally in carrying out the routine accurately. When it is considered that 
he was traveling rapidly and while on the steamer was obliged to make 
all his observations and measurements in the narrow confines of a state- 
room having three other occupants, it will be seen that it is much to 
his credit that the records were so carefully kept. Although it was im- 
possible to keep an accurate record of the amount of food eaten, and 
particularly the kind and amount of the various proteins, a study of 
these urine samples should give some information as to the normal con- 
sumption of protein by this individual. 

From the time of his arrival at the Nutrition Laboratory, the col- 
lection, measurement, and sampling of the urine were made by members 
of the laboratory staff. Particular attention was given to the urine 
excreted during the fasting period, as it was especially important to 
study the entire output of the body at this time. 

When the preliminary arrangements were made for the analyses and 
their assignment to the various members of the laboratory staff and its 
co-workers, it soon became apparent that the number of determinations 
necessary would require a greater volume of urine than would ordinarily 
be passed by a fasting man. It was therefore arranged, in accordance 
with a suggestion made by Dr. Cathcart, to provide the subject with a 
liberal and constant supply of drinking-water. Furthermore, the 
smallest volume of sample which would give accurate determinations 
was carefully considered in order to obtain the greatest number of 
results with the available material. Had it not been for the recent 
development of the new Folin methods, it is probable that much 
valuable data would have been lost. For example, while formerly 
300 c.c. or even more were required for the determination of the uric 
acid, with the new Folin method 5 c.c. would suffice. Many of the 
determinations of the ammonia as carried out by the new method were 
also made with a relatively small amount of urine. 

Before the subject came under observation the time of urinating was 
more or less irregular. During the three food days preceding the fast, 
the subject urinated at irregular times, although ending each day at 
approximately 8 a. m. During the fasting period, he was required 
to empty the bladder immediately after coming out of the bed calori- 
meter in the morning, this being usually not far from 8 o'clock. He 
again emptied the bladder shortly before entering the bed calorimeter 
at night. We were thus able to divide the urine into two periods, each 
approximately 12 hours in length. Use was made of this routine in the 
latter part of the fast to study the apportionment of the nitrogen and 
ammonia excretion between the day and night periods. 



238 



A STUDY OF PROLONGED FASTING. 



The urine was collected at the laboratory by having the subject 
urinate into a previously dried and weighed bottle; the bottle and con- 
tents were then carefully weighed and the urine measured in a gradu- 
ate and the volume recorded. Shortly after the experiment began, 
it was considered advisable, in accordance with a suggestion made by 
Dr. Folin, to add sufficient distilled water to bring the urine to a definite 
volume each day. Under these circumstances a normal excretion of 
urine of 600 to 700 c.c. would be weighed and its specific gravity deter- 
mined; it would then be immediately diluted to 1,000 c.c. and division 
made for the various analyses. This procedure was very satisfactory 
and minimized the calculations. 



COMPOSITION OF THE URINE PRIOR TO THE FASTING EXPERIMENT. 

As a general indication of the character of the subject's urine prior 
to the fasting experiment, we have fragmentary data regarding the 
urine passed on the 10 days before he arrived at the laboratory and for 
the 3 food days in Boston before the fast- 

ing period began. The volume and nitro- ^^tlTS'""''™ 
gen content of this urine are given in table 
23, the nitrogen being determined by the 
Kjeldahl method. In addition to the 
tabulated data, the ammonia-nitrogen 
was determined for the last 3 days by 
the old Folin method, the amounts found 
being 0.67, 0.65, and 0.59 gram respect- 
ively. On the last 2 days the heat of 
combustion was 129 and 104 calories 
respectively; the total carbon in the urine 
for the same days was 11.41 and 9.08 
grams respectively. The acidity was 
determined on but one day (April 11-12), 
this, expressed as cubic centimeters of 
N/10 NaOH solution, being 409 c.c. 

These data will be used in subsequent discussions and are here recorded 
to avoid confusion with the regular examinations of urine in connection 
with the fasting experiment. 

PHYSICAL CHARACTERISTICS OF THE FASTING URINE. 

In considering a subject as complex as is the urinary excretion, it is 
advantageous to note first the physical characteristics and then the 
chemical composition. The influence of various physical agencies, 
particularly the relation between the amount of water drunk and the 
volume of urine, may not be without influence upon the chemical 
composition, for under certain conditions there may well be a washing 



Date. 


Volume 


Nitrogen 




of urine. 


in urine. 


1912. 


c.c. 


grams. 


Apr. 1-2 


1,095 


12.07 


2-3.... 


975 


10.90 


3-4. . . . 


1,208 


16.03 


4-5. . . . 


608 


8.83 


5-6. . . . 


581 


11.90 


6-7. . . . 


818 


10.45 


7-8. . . . 


795 


11.30 


8-9.... 


1,215 


13.36 


9-10. . . 


1,151 


12.25 


10-11... 


1,485 


17.02 


11-12. . . 


1,521 


15.92 


12-13... 


1,528 


14.48 


13-14. . . 


1,441 


11.54 



URINE. 



239 



out of the end-products of protein katabolism by the excess water. 
Furthermore, the specific gravity (when accurately determined) and 
also the total solid matter have an interest second only to the chemical 
constituents of the urine. Accordingly in table 24 a record is given 
of the water consumed, the volume of urine, the amount of urine in 
grams, the water in the urine, the ratio of water in the urine to the 
water consumed, the specific gravity, the total solids, either computed 
or determined, and the ratio of the total solids to the specific gravity. 



Table 24. 



-Relations between water consumed, water in urine, specific gravity, and total solid 
matter in experiment with L. 















a hi • 




a 






Day of 


73 

« 

s 

■ 
a 
o 


6 
a 

1 

"o 


6 

a 

1 

O 


6 

a 
°E 

3 

a 


Ratio of water i 
urine to wate 
consumed (d-*-a' 




T3 . 

o a 

CO -g 


o '3 

a, 

B 

°3 .• 


Date. 


fast. 


6 
fa 

8 
■*» 


a 

j3 

> 


43 

'3 




e 
«3 

"3 
9 

do 


3 
O 

H 


Ratio 
solids 
gravit} 






A 


B 


C 


D 


E 


F 


G 


H 


1912 




gm. 


ex. 


gm. 


gm. 






gm. 




Apr. 14-15 


1st 


720 


660 


673.7 


630.2 


0.875 


1.0206 


2 43.51 




15-16 


2d 


750 


468 


482.0 


436.6 


.582 


1.0303 


2 45.38 






16-17 


3d 


750 


565 


581.2 


530.6 


.707 


1.028 


2 50.62 






17-18 


4th 


750 


713 


730.5 


674.4 


.899 


1.0246 


2 56.13 






18-19 


5th 


750 


667 


682.6 


633.5 


.845 


1.023 


2 49.09 






19-20 


6th 


750 


610 


623.9 


577.8 


.770 


1.0236 


2 46.07 






20-21 


7th ... . 


750 


524 


536.5 


495.9 


.661 


1.0242 


2 40.58 






21-22 


8th ... . 


750 


587 


601.0 


556.9 


.743 


1.0235 


2 44.14 






22-23 


9th.... 


750 


607 


622.1 


575.3 


.767 


1.0241 


2 46.81 






23-24 


10th.... 


750 


565 


577.8 


535.3 


.714 


1.0235 


2 42.49 






24-25 


11th.... 


900 


564 


577.2 


535.1 


.595 


1.0233 


2 42.05 






25-26 


12th.... 


900 


517 


529.1 


489.9 


.544 


1.0237 


2 39.21 






26-27 


13th.... 


900 


561 


574.3 


532.3 


.591 


1.0234 


2 42.01 






27-28 


14th.... 


900 


647 


659.9 


619.3 


.688 


1.0196 


2 40.58 






28-29 


15th 


900 


758 


768.3 


735.8 


.818 


1.0134 


2 32.50 






29-30 


16th 


900 


889 


902.3 


861.2 


.957 


1.0154 


41.12 


3.0 


Apr. 30-May I... 


17th.... 


900 


848 


860.9 


821.4 


.913 


1.0153 


39.51 


3.0 


May 1-2 


18th 


900 


657 


668.8 


633.3 


.704 


1.0177 


35.47 


3.1 


2-3 


19th 


900 


728 


739.6 


705.0 


.783 


1.0153 


34.59 


3.1 


3-4 


20th 


900 


699 


708.7 


678.6 


.754 


1.0143 


30.06 


3.0 


4-5 


21st 


900 


708 


717.2 


685.3 


.761 


1.013 


31.88 


3.5 


5-6 


22d 


900 


785 


794.6 


763.4 


.848 


1.0127 


31.18 


3.1 


6- 7 


23d 


900 


556 


565.8 


536.5 


.596 


1.0176 


29.30 


3.0 


7-8 


24th 


900 


750 


759.5 


727.5 


.808 


1.013 


32.01 


3.3 


8-9 


25th 


900 


713 


722.1 


691.8 


.769 


1.0135 


30.32 


3.1 


9-10 


26th 


900 


728 


737.1 


706.1 


.785 


1.0129 


31.04 


3.3 


10-11 


27th 


900 


653 


662.7 


631.2 


.701 


1.0147 


31.52 


3.3 


11-12 


28th 


900 


655 


663.4 


634.3 


.705 


1.0134 


29.06 


3.3 


12-13 


29th 


900 


697 


705.8 


676.2 


.751 


1.0129 


29.64 


3.3 


13-14 


30th 


900 


771 


780.3 


750.7 


.834 


1.0119 


29.58 


3.2 


14-15 


31st 


900 


566 


574.5 


547.4 


.608 


1.0150 


27.07 


3.2 



lf The amounts of water in urine from Apr. 14-15 to Apr. 28-29 have been obtained by means 
of computed amounts of total solids. 

2 Calculated by means of the average ratio (3.2) of total solids to specific gravity, determined 
in the last 16 days of the fast. For the formula U3ed in the computation, see Benedict, Carnegie 
Inst. Wash. Pub. No. 77, 1907. page 354. 



240 A STUDY OF PROLONGED FASTING. 

VOLUME OF URINE. 

In the publication giving the results of the earlier fasting research, 
it was clearly brought out that no one factor affects the volume of 
urine as does the volume of the water ingested, particularly when the 
volume of drinking-water is over 1,000 c.c. In the experiment with 
L., the water consumed was always under 1 liter and hence the influence 
of the amount of the drinking-water on the volume of the urine was not 
so obvious. 

The volume of urine varied from 468 c.c. on April 15-16 to 889 c.c. 
on April 29-30. The average volume was 659 c.c. On April 24-25, 
the volume of the water taken was changed from 750 c.c. to 900 c.c, 
and it was expected that this change would materially affect the volume 
of the urine. An inspection of the data shows, however, that it pro- 
duced no marked effect upon the water excreted in the urine, at least 
during the first 3 days. Thus, on the 9 days from the second to the 
tenth days of the fast, inclusive, when the subject drank 750 c.c. of 
water daily, the average volume of urine was 590 c.c. per day and on the 
3 days from the eleventh to the thirteenth days of the fast, inclusive, 
when the daily amount had been increased to 900 c.c, the average 
volume of urine per day was 547 c.c. Subsequently there was a dis- 
tinct tendency for the urine volume to increase and on the next 10 days 
the average volume was 728 c.c, an increase of 138 c.c. over the period 
when 750 c.c. of water was taken, closely approximating the increase in 
the amount of drinking-water. It is to be noted, however, that at this 
stage of the fast, 900 c.c. of drinking-water was proportionately large 
for this man's needs, since he had decreased materially in weight. In 
the earlier fasting experiments, when several liters of water were taken 
daily, the amount of water drunk unquestionably influenced the 
amount of water in the urine excreted, but with the comparatively 
small amount taken by the subject L., the effect was evidently at a 
minimum, and the absence of any flushing-out of the end-products of 
protein katabolism simplifies the subsequent discussion. 

In comparing the water drunk with the volume of urine, the discus- 
sion may be based more advantageously upon the water in the urine. 
A determination of the solids in the urine was made only on the last 
16 days of the fast, but amounts for the earlier days of the fast have 
been computed and these are sufficiently accurate to use in this con- 
nection for obtaining the amount of water in the urine. 

From the values given in column e of table 24, showing the relation- 
ship between the water in the urine and the water consumed, it is seen 
that in the first days of fasting, when the amount of water taken by 
the subject was only 750 c.c, about 74 per cent of the water consumed 
appeared in the urine. When the amount of drinking-water was 
increased to 900 c.c, there were marked disturbances in the ratio for 
the next 7 days. The widest variations in the entire series appeared in 



URINE. 241 

these 7 days, namely, from 54 per cent on April 25-26 to as high as 96 
per cent on April 29-30. On the other hand, from May 1-2 to the end 
of the fast, the ratio remained very constant, with but minor variations 
above or below the average figure for the whole series of 0.744, essen- 
tially that obtaining on the first 10 days of the fast. The disturbance 
in the ratio was found, therefore, only during the 7 days immediately 
following the change in the amount of water consumed from 750 c.c. 
to 900 c.c. 

This surprising constancy in the ratio between the water of urine and 
the water consumed, aside from the 7 days mentioned, is difficult to 
explain, particularly since at least two factors might have been ex- 
pected to disturb this relationship. During the first 10 days of the 
fast, there was a considerable loss of preformed water from the body, 1 
ranging from 769 grams to 183 grams. This loss of water might be 
expected to increase the volume of urine during these days. Indeed, if 
the volume of urine were not increased, in the absence of other evidence, 
this might be taken as an argument against such an excretion of pre- 
formed water. 

Another point which should be taken into consideration in this 
connection is that one would expect that with the greatly diminished 
body substance, the amount of drinking-water consumed might exceed 
the physiological need and hence would disturb the relationship 
between the volume of urine and the volume of water consumed. On 
the other hand, it is well known that during fasting there is a tendency 
for all the tissues to become water-rich, this retention of water possibly 
compensating for the decrease in the physiological need following the 
decrease in the size of the organism. 

That the relationship between the water of urine and the water 
consumed is reasonably constant, even when the quantity of water is 
but 1 liter or less, is likewise substantiated by calculations from data 
published regarding Cathcart's experiment with Beauts, in which the 
volume of drinking-water was also constant, i. e., 1,000 c.c. per day. By 
using the data for the volume of urine, the specific gravity, and the 
factor 3.2 (see page 244), we have computed the total solids and also 
the water in the urine for the sixth, seventh, eighth, tenth, eleventh, 
twelfth, and fourteenth days of Beaute's fast. The ratio of water in 
the urine to the water consumed was for the several days as follows: 
0.813; 0.649; 0.630; 0.940; 0.612; 0.576; 0.662. The average value 
was 0.697, which is not materially different from 0.750, the average 
obtained for the values for L. when the 7 days referred to have been 
omitted. The high value of 0.940 found on the tenth day with Cath- 
cart's subject exceeds any found with our subject L. 

In general, then, the volume of water in the urine is approximately 
75 per cent of the amount of drinking-water taken, even when but 

1 See discussion of preformed water loss on page 408. 



242 A STUDY OF PROLONGED FASTING. 

1 liter or less is taken, provided the intake of water is constant. It 
is obvious, however, that this would hold true only when the factors 
influencing the loss of water, such as environmental temperature and 
exercise, also remain constant. 

The evidence is clear, therefore, that even with the small amounts of 
water taken in this fast, there was a reasonably constant relationship 
between the water consumed and the water in the urine, the absolute 
fluctuations in the volume of urine noted being so small that there 
could have been no disproportionate washing out from the tissues of 
the crystalline end-products of protein katabolism. There is, to be 
sure, a distinct increase in the average volume of the urine after the 
twelfth day, but in this experiment we deal with an average increase 
of approximately 100 c.c, and hence these absolute variations in volume 
are not to be compared with the very large variations noted in the 
earlier fasts at Wesleyan University, when the amount of drinking- 
water varied within wide limits. 

SPECIFIC GRAVITY. 

The specific gravity of the urine was carefully determined by Miss 
Alice Johnson, at a constant temperature of 20°C, on a Westphal bal- 
ance. The position of the scale when the weight was suspended in 
distilled water was accurately checked, the temperature of the sur- 
rounding liquid being invariably artificially maintained at 20° C. 
Thus the specific gravity of the urine for each day of the fast was 
readily obtained to the fifth significant figure. 

These values, which are given in table 24, ranged from 1.0303 on 
April 15-16 to 1.0119 on May 13-14, and are therefore well within 
normal limits. As the fast progressed, there was a distinct tendency 
for the specific gravity to decrease, although for the first 10 or 12 days 
it was approximately constant at about 1.024, falling thereafter some- 
what sharply and remaining at about 1.015 for the remainder of the fast. 
This approximation to constancy may be in part accounted for by the 
approximately constant volumes of urine passed. In the short fasting 
experiments at Wesleyan University, the specific gravity ranged be- 
tween 1.0338 and 1.0032, but the lowest specific gravity was accom- 
panied by a very large volume of urine and the high specific gravity 
by a very small volume of urine. 

It is obvious that the nature of the solids dissolved in the urine has 
a noticeable influence upon the specific gravity. For instance, a solu- 
tion containing 100 grams of sodium chloride in 1 liter has a density at 
15° C. of 1.073, while a solution of urea 1 to 10 has a density of but 
1.028; consequently a large amount of sodium chloride in the urine 
would have considerable effect upon the specific gravity. From other 
fasting studies it is known that a large amount of sodium chloride is 
excreted in the first days of fasting, which would thus increase the 



URINE. 243 

specific gravity. After the first few days, the excretion would be in 
large part of an organic nature, accompanied by the usual salts other 
than sodium chloride; hence we should normally expect to find the 
specific gravity somewhat lower in the latter part of the fast. 

So close is the relationship between the total solids and the specific 
gravity that a formula has been in use for many years for computing 
the total solids by means of this index. Thus, the approximate weight 
in grams of total solids in 1 liter of normal urine may be calculated by 
multiplying the last two figures of the specific gravity (as ordinarily 
expressed in 3 decimal places) by the factor 2.33. The values for the 
specific gravity were so used for computing the total solids in the urine 
for a part of the fasting period, substituting 3.2 as the factor. 

TOTAL SOLIDS. 

In the pressure upon the laboratory staff necessary for carrying out 
the many details of this elaborate research, the determination of the 
total solids in the urine was unfortunately overlooked until the latter 
part of the fast. The data secured in previous researches as to the 
total solids for the first days of fasting are, however, fairly complete, 
and we are thus able to supplement these by the important data which 
were obtained in the last part of this fasting experiment. The proced- 
ure followed in determining the total solids was primarily developed 
for the determination of carbon in urine, and the description of the 
method applies likewise to the method used for securing the carbon 
excretion. Three samples of each specimen of urine were prepared in 
the following manner : 

A small soft-metal bottle-cap was first weighed and in this were 
placed 50 milligrams of pure salicylic acid. With a carefully calibrated 
pipette, 20 c.c. of urine were next added. The bottle-caps containing 
the acid and the urine were then placed on the laboratory table in such 
a position that a current of air from an electric fan would blow over 
them. This drying was continued over night, usually for a period of 
about 24 hours. The samples were next dried for 24 hours in a high 
vacuum in a desiccator. Subsequently the lead capsules with their 
dried contents were quickly weighed, the 50 milligrams of salicylic 
acid being subtracted from the final weight of dry matter, thus giving 
the weight of the total solids in 20 c.c. of urine. 

In determining the carbon and the heat of combustion by this 
method, it is unnecessary to dry the substance in the capsules com- 
pletely in a high vacuum, but at the end of the 24-hour drying in the 
current of air, the thick pasty material may at once be transferred 
to the small capsules used in the combustion bomb. It is thus seen 
that had we only delayed the weighing of the soft-metal bottle-caps for 
24 hours, it would also have been possible for us to determine the total 
solids in the urine for the first part of the fast. 



244 A STUDY OF PROLONGED FASTING. 

When the total amount of dry matter was determined, the contents 
were afterwards carefully transferred from the metal bottle-cap to the 
nickel capsule by means of a swab of ignited asbestos, the last traces 
of solid material adhering to the bottle-cap being removed by a bit of 
asbestos wool moistened with water. Finally, the material was dried 
in an air-current to a pasty consistency and then placed in a high 
vacuum until dry enough to burn. 

This method has been previously described 1 and need only be referred 
to here. The 3 samples always gave perfectly agreeing results, indica- 
ting that the drying was essentially complete, and testifying to the 
skilful technique of Mr. Arthur W. Cornell, who carried out the deter- 
minations of the total solids, carbon, and heat of combustion. The 
results of the determinations of the total solids for the last 16 days of the 
fast are given in table 24, and range from 41.12 grams on April 29-30 
to 27.07 grams on the last day of the fast. From these absolute 
determinations of the total solids, together with the volumes of urine 
and the specific gravity, it was possible to compute a factor indicating 
the ratio between the total solids and the specific gravity. This factor, 
although higher than the value for normal individuals (2.33), remained 
very constant in these later days of the fast, ranging from 3.0 to 3.5, 
with an average of 3.2. The value 3.2, which represents the average 
ratio between the total solids and the specific gravity in the last part 
of the fast, agrees very closely with those found in three of the fasting 
experiments with the subject S. A. B. in the Middletown research, 2 
namely, experiments Nos. 71, 73, and 75, the ratios being 3.0, 3.4, and 
3.3 respectively. In experiment No. 77 with the same subject, much 
larger amounts of sodium chloride were excreted, and this doubtless 
was the cause, at least in part, for the lower ratio of 2.5 for this 4-day 
fasting experiment. In the later days of the fasting experiment with 
our subject L. there was undoubtedly a minimum sodium-chloride 
excretion, and the constancy in the ratio between the total solids and 
the specific gravity points towards an approximately constant relation- 
ship between the organic and inorganic solids of the urine. Unfor- 
tunately, we are unable to apportion the total solids between the min- 
eral and organic constituents, since it was impossible to determine 
the ash content of the fasting urine, owing to the deficiency in material. 
It can only be pointed out here, therefore, that while the factor 3.2 
is considerably larger than that accepted for normal people, namely, 
2.33, it is probably explained in part by the fact that there were pro- 
ducts of defective fat katabolism in the urine. 

This average ratio, i. e., 3.2, was used for computing the total solids 
for the first 15 days of the fast. The results of these computations are 
also given in table 24. In comparing the values we find that the largest 

1 Higgins and Benedict, Am. Journ. Physiol., 1911, 28, p. 291. 
Benedict, Carnegie Inst. Wash. Pub. 77, 1907, p. 355. 



URINE. 



245 



amount was 56.13 grams on April 17-18. It is probable that, owing 
to the increased sodium-chloride excretion in the first days of fasting, 
the values are somewhat too high and that the factor used should have 
been less than 3.2. 

DAY AND NIGHT URINES. 

While with a fast as prolonged as this the main interest lies in a 
comparison of the urinary excretion from day to day and there is but 
little interest in a subdivision of the day into 12-hour periods, yet we 
have certain fragmentary data regarding the diurnal excretion of urine 
which are of sufficient importance to record here. Usually the bladder 
was emptied at 8 a. m. and at 8 p. m., the 24-hour day being thus divided 
into two periods, i.e., from 8 a. m. to 8 p. m. and from 8 p. m. to 8 a. m. 
While this subdivision was not made exactly each day, nevertheless 
the variations were generally well within one-half hour. On one day 

Table 25. — Periodic distribution of volume and nitrogen of urine in experiment with L. 







Day period. 


Night period. 








Nitrogen. 








Nitrogen. 


Date. 


Day 

of 

fast. 




































Duration. 


Vol. 




Propor- 


Duration. 


Vol. 




Propor- 










A 


mt. 


tion of 






A 


mt. 


tion of 










il 


total for 










total for 














24 hours. 










24 hours. 


1912. 




a.m. 


p.m. 


c.c. g 


m. 


p. ct. 


p.m. 


a.m. 


c.c. g 


m. 


p. ct. 


Apr. 18-19.... 


5th. 


8 h 05 m 


to 7 h 27 m 


307 . 






yb 27m 


to 7 h 57 m 


360 . 






19-20 


6th. 


7 57 


8 06 


312 . 










8 06 


8 05 


298 . 






20-21 


7th. 


8 05 


8 02 


256 . 










8 02 


8 04 


268 . 






21-22 


8th. 


8 04 


8 05 


298 . 










8 05 


7 59 


289 . 






22-23 


9th. 


7 59 


8 05 


315 . 










8 05 


8 05 


292 . 






23-24 


10th. 


8 05 


8 03 


305 . 










8 03 


7 57 


260 






24-25 


11th. 


7 57 


8 08 


284 . 










8 08 


8 04 


280 . 






25-26 


12th. 


8 04 


8 05 


267 . 










8 05 


7 55 


250 . 






26-27 


13th. 


7 55 


7 22 


288 . 










7 22 


8 14 


273 . 






27-28.... 


14th. 


8 14 


7 25 


252 . 










7 25 


8 07 


395 . 






28-29 


15th. 


8 07 


7 00 


263 . 










7 00 


7 55 


495 . 






29-30 


16th. 


7 55 


6 55 


407 . 










6 55 


7 54 


482 . 






Apr. 30-May 1 


17th. 


7 54 


6 53 


328 . 










6 53 


8 06 


520 . 






May 1-2 


18th. 


8 06 


6 48 


284 . 










6 48 


8 00 


373 . 






2-3.... 


19th. 


8 00 


6 45 


250 . 










6 45 


8 07 


478 . 






3-4.... 


20th 


8 07 


6 45 


251 










6 45 


8 01 


448 . 






4-5.... 


21st . 


8 01 


8 02 


'300 3 


.71 


i 


to 


8 


8 02 


8 10 


x 418 4 


.22 


53 .2 


5-6.... 


22d.. 


8 10 


8 02 


335 










8 02 


7 57 


450 . 






6-7.... 


23d.. 


7 57 


8 11 


266 . 










8 11 


8 00 


290 . 






7-8.... 


24th. 


8 00 


8 00 


»316 3 


.94 


47 


8 


8 00 


7 55 


!443 4 


!30 


52.2 


8-9.... 


25th. 


7 55 


8 07 


323 3 


.84 


48 


5 


8 07 


8 05 


390 4 


.07 


51.6 


9-10.... 


26th. 


8 05 


8 12 


282 3 


.96 


49 


5 


8 12 


8 10 


446 4 


.04 


50.5 


11-12 


28th. 


8 08 


8 08 


242 3 


.34 


44 


8 


8 08 


7 39 


413 4 


.12 


55.2 


12-13 


29th. 


7 39 


8 05 


300 3 


.82 


50 





8 05 


7 41 


396 3 


.82 


50.0 


13-14 


30th. 


7 41 


8 00 


311 3 


.74 


49 


2 


8 00 


7 44 


460 3 


.86 


50.8 


14-15 


31st . 


7 44 


8 09 


266 3 


.30 


48 


3 


8 09 


8 02 


300 3 


.53 


51.7 



KJra 



246 A STUDY OF PROLONGED FASTING. 

(May 10-11) the day period was 14 hours and 5 minutes and the night 
period only 9 hours and 55 minutes, a discrepancy which resulted in the 
omission of the day from the record of the periodic distribution of the 
urine. Beginning with April 18-19, the volumes of the day and night 
urines were separately recorded. Furthermore, toward the end of the 
fast, periodic determinations of the nitrogen for the day and night 
were made. The data thus obtained are given in table 25. Without 
laying emphasis at this time on the absolute values of total nitrogen 
excreted — a discussion which belongs later in this report — we may 
properly consider the data in this table as indicating the division of 
the urinary excretion between night and day when no food was taken. 

The volume of urine during the day ranged from a minimum of 
242 c.c. on May 11-12 to a maximum of 407 c.c. on April 29-30. The 
average volume for the day urine was 293 c.c. The average volume for 
the night urine was 376 c.c, an increase of 83 c.c. On several occasions 
there were large differences between the day and night urines, which 
are not easily explained. Thus, on April 28-29 there were but 263 
c.c. in the daytime and 495 c.c. during the night. Although the day 
period was but 11 hours and the night period 13 hours, this difference 
in volume is very large. While in general the volumes for the day and 
the night were not far apart, the average difference, as we have seen, 
being but 83 c.c, it is indeed surprising that a larger volume was not 
excreted during the day, for the subject drank his entire allotment 
of 900 c.c. of water before 8 p. m., taking it in fairly regular portions 
throughout the day. Usually but a small amount was left after 6 p. m. 
The drinking of water was thus distributed to obviate the necessity for 
urinating inside the calorimeter chamber during the night; and indeed, 
throughout the fast, the subject retained the urine in the bladder 
the entire night period. 

On the days for which we have the data for both the volume and 
the nitrogen of the urine, we find that on an average 42 per cent of the 
total volume of urine and 48.1 per cent of the total nitrogen were 
excreted during the daytime. 1 If the volume of urine had had a 
material effect upon the total nitrogen, one would expect that a some- 
what greater proportion of nitrogen would have been excreted during 
the night than was actually found, and it is reasonable to suppose that, 
with the relatively small total volume of urine here involved there could 
have been but little washing out of the nitrogenous products as a result 
of the differences in volume. But in view of the well-known fact that 
large quantities of water assist in washing out nitrogenous material, no 
other explanation than the increase in the volume of urine seems possi- 
ble for this small but positive increase in the nitrogen output during 
the night. 

1 Obviously slight corrections for variations in the relative length of the day and night periods 
should be made, but an inspection of the table shows that the percentage figures would not be 
materially altered. 



URINE. 247 

CHEMICAL CONSTITUENTS OF FASTING URINE. 

In modern urinary analysis, as carried out in connection with 
a metabolism experiment, we have several distinct classifications or 
subdivisions : First, the total nitrogen and the partition of nitrogen in 
accordance with the analytical scheme of Folin; second, the acid 
radicles, which would include the chlorine, phosphorus, sulphur, total 
acidity, and /3-oxybutyric acid; third, the bases — calcium, magnesium, 
potassium, and sodium oxides —whose excretion might perhaps be 
discussed in connection with the ammonia, itself a base; and finally, 
attention should be given to the determination of the reducing power, 
total carbon, and the heat of combustion of urine. 

With these various determinations, several ratios can be intelligently 
discussed, of which the most important may be the ratio of nitrogen 
to sulphur, nitrogen to phosphorus, carbon and energy to nitrogen, 
and the heat of combustion to carbon. The presentation of the data 
secured in our fasting experiment with L. will follow essentially the 
analytical scheme thus outlined. 

TOTAL NITROGEN. 

From the early days of the Liebig titration method for determining 
total nitrogen down through the various modifications to the present-day 
development of the Kjeldahl technique, one of the first and most impor- 
tant determinations of the constituents of the urine in metabolism experi- 
ments has been that of the total nitrogen, the importance increasing in 
proportion as the technique has been developed. After the accuracy of 
the Kjeldahl method had been demonstrated and an exact method 
was thus available, we were informed by Folin, as a result of his beauti- 
ful systematic analyses of the urine in which the partition of the nitro- 
gen has been made, that the value of the total nitrogen in the urine 
did not have the importance which had formerly been attributed to it. 
For instance, the determination of the carbon-dioxide excretion in a 
metabolism experiment has great value in itself, but the apportionment 
of this carbon dioxide to fat, carbohydrate, and protein katabolism 
has a much greater value; similarly, although the determination of the 
total nitrogen in the urine is not without value, yet the apportionment 
of the nitrogen among the various constituents of the urine is much 
more illuminating and scientifically intelligible than the amount of 
the total nitrogen. The determinations of the total nitrogen in the 
fasting urine were therefore made primarily as a preliminary to con- 
sidering the partition of the nitrogen. 

Comparison of Total Nitrogen Excretion of L. with that of Other Fasting Subjects. 

While the total nitrogen has been determined in the greater number 
of prolonged fasting experiments, in relatively few of these studies has 
the determination been made by the modern Kjeldahl method, the 



248 A STUDY OF PROLONGED FASTING. 

results in many experiments having been recorded in terms of "urea." 
Nevertheless, the nitrogen values found in several fasts of 7 or more 
days are considered of sufficient importance in connection with the 
study of the fasting urine of our subject to be reproduced here and 
are accordingly recorded in table 26. 

In this table the body-weight at the beginning of the fast is given 
for nearly every subject, and frequently for comparison the nitrogen 
excretion is included for the day prior to the fasting. The values for 
L. are first shown, these being followed by the nitrogen found in Succi's 
fasts. Unfortunately these latter values are not strictly comparable 
with the others, owing to the differences in methods of determination. 
Those found for Cetti can be relied upon, as can those for Beauty, 
Schenk, Tosca, and S. A. B. The values reported for Succi for the 
London and Naples fasts are undoubtedly somewhat low, but those 
for the Florence fast have been corrected by Munk. Even when these 
points are taken into consideration, the most striking feature in this 
whole group of results is the fact that the nitrogen excretion of our 
subject L. continues to be extraordinarily high to the fifteenth day of 
the fast, and, indeed, throughout the remainder of the fast the values 
are noticeably higher than those found in any other study of prolonged 
fasting. 1 Values as high, and even higher, are shown for Cetti for 
the 10 days of his fast, and also for the 7-day experiment of S. A. B., 
but in none of the longer fasts are such high values so continuously 
shown. In Succi's 30-day fast in Hamburg the value found for the 
last day (8.42 grams) was higher than that for the thirtieth day of 
the fasting experiment with L., but the earlier values were measurably 
lower. Another point of interest is that the general tendency is toward 
a low nitrogen output on the first day of the fast, with a higher nitrogen 
excretion on the subsequent one or two days. This characteristic is 
shown in the fasts with L., S. A. B., Tosca, and Beauts, and may 
easily be attributed to the protecting action of the body-storage of 
glycogen during the first few days. 

One striking fact in connection with the high nitrogen output in 
L.'s fast is that on the fifteenth day there was a sudden fall of nearly 
2 grams. An inspection of the values for the other subjects shows 
that in all of the fasts this sudden fall in the nitrogen excretion occurred 
at some point. Thus, in Succi's fast at Florence, there was a fall of 
1 gram on the eighth day; in the London fast there was a fall of 1.2 

x None of these subjects shows as absolute a minimum value for nitrogen excretion as was found 
on one day with Grafe's insane subject (Grafe, Zeitschr, f. physiol. Chemie, 1910, 65, p. 21), when 
the very low excretion of 1.057 grams of nitrogen was found. Since the body-weight of this 
subject was at the time 49.25 kilograms, this excretion would correspond to approximately 0.0215 
gram of nitrogen per kilogram of body-weight. This surprisingly low value is difficult to explain, 
for while Grafe states that during the latter part of the experiment the urine was frequently spon- 
taneously passed and hence the 24-hour periods could not be accurately determined, yet his dis- 
cussion of this low value of 1.057 grams indicates that he believed it represented a 24-hour excre- 
tion of nitrogen. This remains the lowest value that we have as yet seen reported in any fasting 
observation on men or women. 



URINE. 



249 



grams on the seventh day; in the Naples fast, the nitrogen output 
fell 1.8 grams on the eighth day; in the Rome fast, it fell 1.8 grams on 
the ninth day; while in the Vienna fast, it fell 2.9 grams on the tenth 
day. With Cetti there was a fall in the nitrogen excretion of 2 grams 
on the eighth day; with Beaute* the decrease was 1.1 grams on the 
seventh day; with Schenk it was 1 gram on the tenth day; with Tosca, 





Table 26 - 


-Nitrogen eliminated in urine 


daily by fasting subjects 








Day of fast. 


co 

5 

M 

CO 

d 

CO 


Succi. 


o5 
o 
r3 
M 

© 

'43 
o 
O 


CO 

O 

% 
CO 

ui 

CO 
WU 

3 
03 

8 

PQ 


m 

o 

M 

eo 

CO 
"5 

^5 

a 

<u 
A 

a 
GO 


03 
8 

CO 

o 


GO 

o 

IQ 
OJ 

»o 

PQ 

< 

03 


■ CO 

<u o 

5 -- 

r 1 CO 
fo CO 


00 

a oo 


A 

o 
co rS 

a ja 

Pi CO 

i< CD 


CO 

O 

aj '£ 


c3 

a 
a 

CO 

> 


a in 


Last food day . 
1st 


gm. 

11.54 

7.10 

8.40 

11.34 

11.87 

10.41 

10.18 

9.79 

10.27 

10.74 

10.05 

10.25 

10.13 

10.35 

10.43 

8.46 

9.58 

8.81 

8.27 

8.37 

7.69 

7.93 

7.75 

7.31 

8.15 

7.81 

7.88 

8.07 

7.62 

7.54 

7.83 

6.94 


gm. 

1 17.85 

15.19 

12.13 

15.25 

14.08 

14.12 

11.13 

10.31 

9.37 

8.56 

7.43 

8.67 

7.88 

3.86 

5.87 

5.66 

6.05 

6.78 

6.00 

5.54 

4.82 

4.27 

3.52 

5.23 

6.11 

6.65 

5.57 

5.90 

6.16 

4.49 

7.28 


gm. 

3 11.41 
12.62 
12.00 
10.46 
9.80 
8.57 
7.86 
7.50 
7.10 
6.51 
6.86 
6.14 
5.95 
5.36 
5.35 
5.40 
4.13 
3.98 
4.64 
4.00 
5.16 
4.66 
4.70 
4.32 
4.12 
3.81 
3.40 
4.13 
4.56 
4.36 
4.79 
4.75 
4.80 
4.80 
3.95 
5.00 
4.77 
4.99 
5.56 
5.82 


gm. 
2 8.99 
8.72 
8.45 
9.05 
8.51 
9.87 
8.62 
7.62 
5.84 
6.90 
5.37 
5.10 
6.19 
4.83 
3.83 
4.14 
3.24 
5.01 
4.06 
3.49 
4.77 
5.37 


gm. 
9.13 
8.91 
9.17 
8.68 
8.46 
10.01 
9.42 
8.58 
8.14 
6.35 
5.71 
4.94 
5.11 
4.78 
4.41 
2.83 
3.15 
3.32 
4.06 
3.82 
3.45 


gm. 


gm. 


gm. 
13.49 
13.55 
12.59 
13.12 
12.39 
10.70 
10.10 
10.89 

8.90 
10.83 

9.47 


gm. 

16.45 

10.51 

14.38 

13.72 

13.72 

(11.30) 

10.77 

9.67 

9.52 

(9.39) 

8.38 

8.49 

8.77 

(8.97) 

7.78 


gm. 

8.41 
6.50 
7.78 
7.86 
7.82 
7.13 
6.20 
5.40 
4.38 
5.17 
5.38 
8.11 
5.96 
5.10 
4.07 


gm. 

13.99 
8.76 
8.38 

10.73 
9.40 
7.87 
7.73 
6.11 
7.70 
7.35 
6.80 
6.14 
6.97 
5.62 
4.08 


gm. 
19.50 
12.24 
12.45 
13.02 
11.63 
10.87 
10.74 
10.13 


17.0 

11.2 

10.55 

10.8 

11.19 

11.01 

8.79 

9.74 

10.05 

7.12 

6.32 

6.84 

5.14 

4.66 

5.05 

4.23 

5.4 

3.6 

5.7 

3.3 

2.82 




2d 


3d 


4th 


5th 


6th 


7th 


8th 


9th 


10th 


11th 


12th 






13th 






14th 






15th 






16th 












17th 












18th 














19th 














20th 














21st 














22d 














23d 








5.84 
6.41 
6.27 
6.18 
6.30 
4.44 
4.19 
8.42 












24th 


















25th 


















26th 


















27th 


















28th 


















29th 


















30th 


















31st 


















32d 




















33d 
























34th 
























34th 
























35th 
























36th 
























37th 
























*8th 
























59th 
























K)th 

















































J The figures in this column are given for the first 10 days of the fast as corrected by Munk. The results for 
t e remaining days have been increased in like proportion. 

*Given by Ajello and Solaro as urea and here converted to nitrogen for purposes of comparison. Since the 
a thors do not give the method employed, no attempt is here made to correct the figures. 

3 The results in this column were reported by the investigators as grains of urea, but are here converted to 
g ams of nitrogen in urea for purposes of comparison. 



250 A STUDY OF PROLONGED FASTING. 

1.5 grams on the fourteenth day; and with S. A. B., 1.4 grams on the 
fourth day. These sudden drops were almost invariably permanent 
and were sometimes followed by a day on which even lower values were 
found. It is difficult to predict at what point this break in the nitrogen 
curve is likely to appear, and the irregularity of certain curves does not 
justify giving serious attention at present to this feature of the general 
course ; nevertheless the fact that it is characteristic of all long fasting 
experiments is worthy of note. 

The most accurate nitrogen determinations for the prolonged fasts 
shown in table 26 are unquestionably those made by Brugsch for Succi's 
fast at Hamburg. These values are somewhat lower than those found 
for L., although Succi's body-weight was 18 kilograms greater than 
that of our subject. In none of the fasting experiments do we find, 
save perhaps in the Hamburg fast, any indication of an increase in the 
nitrogen excretion near the end of the fast which may be considered 
as corresponding to the so-called "pre-mortal" rise which has been 
observed with many fasting animals, particularly with rabbits. It may 
be said, therefore, that the values found for L. follow much the same 
general course as the values found with the subjects of earlier fasting 
experiments, except that the level of the nitrogen excretion after the 
first 7 days was distinctly higher than with the other subjects. 

Daily Excretion or Nitrogen. 

Since L. had carefully preserved specimens of the urine from April 1 
until the time of his arrival at the Nutrition Laboratory, we were able 
to obtain information as to the nitrogen excretion of this subject for 
13 days preceding the fasting experiment. By reference to the results 
of these determinations (see table 23, page 238), it will be seen that in 
general the nitrogen excretion was on a moderately high level, aver- 
aging not far from 13 grams per day, and even exceeding this when the 
low value of 8.83 grams is excluded. 

On his first day in Boston (April 10-11), the nitrogen excretion was 
17.02 grams. This was the highest value found and doubtless resulted 
in part from the large beefsteak eaten by the subject on the night of 
his arrival. The nitrogen excretion subsequently decreased until on 
the last day before the fast it was but 11.54 grams. From April 10 
until the beginning of the fast, therefore, the total nitrogen in the urine 
averaged over 14 grams per day. This is significant as indicating that 
L. was subsisting on a nitrogenous diet, which was quite inconsistent 
with his claim that he was a "low-proteid vegetarian." 

The values for the nitrogen excretion for the whole experiment, 
including not only those for the fasting period, but for the food days 
prior to and following the fast, are given in table 27. 

As noted in the comparison with other fasting subjects, two striking 
features of these values for the total nitrogen excretion are the immedi- 



URINE. 



251 



ate decrease with the beginning of the fast and the continuance of 
the high values until after the fourteenth day. This decrease in the 
nitrogen excretion in the first few days of the fasting period has already 
been explained as being due to the relatively large katabolism of glyco- 
gen on those days. The average nitrogen excretion for the first 10 
days of the fast was over 10 grams. The highest value (11.87 grams) 
in the whole of the series was found on the fourth day, and the lowest 
value (6.94 grams) on the thirty-first day. That the lowest value was 

Table 27. — Nitrogen excreted in urine, per day and per kilogram of body-weight, 

in experiment with L. 



Date. 


Day of 

fast. 


Nitrogen excreted. 


Date. 


Day of 

fast. 


Nitrogen excreted. 


Per day. 


Per 

kilogram 
of body- 
weight 
per day. 


Per day. 


Per 

kilogram 
of body- 
weight 
per day. 


1912. 
Apr. 11-12. . 




gm. 

15.92 

14.48 

11.64 

7.10 

8.40 

11.34 

11.87 

10.41 

10.18 

9.79 

10.27 

10.74 

10.05 

10.25 

10.13 

10.35 

10.43 

8.46 

9.58 


gm. 
0.264 
.238 
.190 
.118 
.142 
.195 
.207 
.184 
.181 
.176 
.186 
.196 
.185 
.190 
.189 
.193 
.196 
.160 
.182 


1912. 
Apr. 30-May 1.. 
May 1-2 

2-3 

3-4 

4-5 

5-6 

6-7 

7-8 

8-9 

9-10 , 
10-11 
11-12 
12-13 
13-14 
14-15 , , 
15-16 


17th... 
18th... 
19th... 
20th . . . 
21st . . . 
22d. . . . 

23d 

24th... 
25th... 
26th... 
27th... 
28th... 
29th... 
30th. .. 
31st. . . 


gm. 
8.81 
8.27 
8.37 
7.69 
7.93 
7.75 
7.31 
8.15 
7.81 
7.88 
8.07 
7.62 
7.54 
7.83 
6.94 
4.83 
3.81 
x 2.75 


gm. 

0.169 
.160 
.163 
.151 
.156 
.154 
.146 
.164 
.158 
.160 
.165 
.157 
.156 
.163 
.146 
.102 
.081 

1 .058 


12-13 . . 




13-14. . 




14-15.. 
15-16.. 
16-17.. 
17-18.. 
18-19 . . 
19-20. . 
20-21 . . 
21-22 . . 
22-23.. 
23-24.. 
24-25.. 
25-26 . . 
26-27.. 
27-28.. 
28-29 . . 
29-30.. 


1st. . . 

2d. . . . 

3d.... 

4th... 

5th .. . 

6th... 

7th... 

8th... 

9th... 
10th... 
11th... 
12th... 
13th... 
14th... 
15th . . . 
16th... 


16-17 




17-18 









determined in urine for about 22 hours. 

but 0.16 gram lower than that found on the first day may be explained 
by the fact that on the first day L.'s energy requirement was in part 
met by the combustion of about 70 grams of glycogen (see table 
63, page 412). On the last day the katabolism was essentially a 
protein-fat katabolism, unassisted by the combustion of any measurable 
amount of carbohydrate. During the 31-day fast this subject actually 
excreted 277.32 grams of nitrogen in the urine, thus averaging 8.95 
grams of nitrogen per day. This would correspond to 1,664 grams of 
protein, or 8,319.60 grams of flesh. Since the entire loss in body-weight 
of this subject was 13.25 kilograms, it can be seen that 63 per cent of 
the total loss may be accounted for in flesh katabolized. 



252 



A STUDY OF PROLONGED FASTING. 



Although this is primarily a study of the excretion of nitrogen during 
fasting, the values found for the 3 days subsequent to the fast have a 
certain interest. During these days the subject took an almost 
protein-free 1 diet, consisting of fruit juices and honey. The large 
amount of carbohydrate contained in this diet immediately protected 
the protein in the body and in consequence there was a continually 
decreasing nitrogen excretion, until on the last day we have the lowest 
amount found with this subject, namely, 2.75 grams. This 22-hour 
value is actually somewhat lower than that found with Beaute" by 
Cathcart in a 3-day experiment with a starch-cream diet of Folin — 
Beaut6, with a body-weight of not far from 58 kilograms, showing a 
minimum nitrogen output of 2.84 grams. Since L. had a body-weight 
at this time of only 47.5 kilograms, it would perhaps be expected that 
his nitrogen excretion would be much lower than that of Cathcart's 
subject; it should be noted, however, that his total nitrogen level was 
considerably higher than that shown by Beaute\ 

The values found for our subject L. on these days of food following 
the fast have a special interest, in that they show that the excess of 
carbohydrate in the diet acted as a great protection of the body protein, 
and hence we have here probably the nearest to the minimum protein 
requirement of this man, corresponding to the "Abnutzungsquote" 
of Rubner. 



Nitrogen Excretion per Kilogram or Body-weight. 

We have no information as to the fluctuations in the body-weight 
prior to the arrival of the subject at the laboratory, but accurate obser- 
vations were made from April 11 to the end of the experiment, and the 
nitrogen per kilogram of body-weight may thus be computed for that 
period. These values are also given in table 27. On the first day of the 
fast the nitrogen output per kilogram of body- weight was very low, being 
only 0. 1 18 gram. It then rose regularly until it reached a maximum on 
the fourth day of 0.207 gram. Thereafter there was, in general, a steady 
fall, with two minima of 0.146 on the twenty- third and the thirty-first 

x To aid in indicating the kinds and amounts of food eaten on the first two days of food fol- 
lowing the fast, the estimated amounts and composition of food eaten are tabulated herewith: 



Date. 


Kind of food. 


Amount 
eaten. 


Protein. 


Fat. 


Carbo- 
hydrates. 


Nitrogen. 


1912. 
May 15-16 

May 16-17 


Lemons 

Oranges 

Honey 

Grape juice. . . 

Total 

Lemon juice . . 

Honey 

Orange juice . . 

Total 


gm. 
100 
450 
311 
1072 


gm. 
1.00 
3.60 
1.24 


gm. 

0.70 

.90 


gm. 
8.5 

52.2 
252.5 
178.6 


gm. 

0.16 
.59 
.19 




5.84 


1.60 


491.8 


0.94 


80 

139 

1128 


6.55 
6.05 




7.8 
112.8 
128.1 


6.08 
1.01 




6.60 




248.7 


1.09 



URINE. 



253 



days respectively. After food was again taken, the nitrogen excretion 
decreased to the surprisingly low value of 0.058 gram per kilogram of 
body-weight. 

Comparison of Methods fob Determining Total Nitrogen and Ammonia-nitrogen. 

The microchemical methods had been developed just previous to 
this fasting experiment and were therefore used by Mr. H. L. Higgins 
for determining the total nitrogen and the ammonia-nitrogen. The 
total nitrogen was also determined by the Kjeldahl method and the 
ammonia-nitrogen by the old Folin method, both determinations being 
made by Miss E. B. Babcock. These analyses of the fasting urines 
were therefore the first control analyses which had been made outside of 
the Folin laboratory. Both the Kjeldahl method and the Folin micro- 
chemical method were frequently tested by determining the nitrogen of 

Table 28. — Comparison of the determinations of nitrogen and ammonia-nitrogen by former 
methods and the new microchemical methods of Folin. 



Date. 


Day of 
fast. 


Total nitrogen. 


Ammonia-N. 


Kjeldahl 
method. 


Folin 


Folin 


Folin 






micro 


method 


micro 






method. 


(original). 


method. 


1912. 




gm. 


gm. 


gm. 


gm. 


Apr. 16-16 


2d 






0.60 


0.60 


16-17 


3d 


11.34 


10.26 


.95 


.95 


17-18 


4th.... 


11.87 


11.46 


1.40 


1.40 


18-19 


5th.... 


10.41 


9.94 


1.60 


1.63 


19-20 


6th.... 


10.18 


9.91 


1.67 


1.69 


20-21 


7th.... 


9.79 


9.30 


1.52 


1.56 


21-22 


8th.... 


10.27 


9.91 


1.62 


1.68 


22-23 


9th 


10.74 


10.74 


1.70 


1.68 


23-24 


10th.... 


10.05 


10.02 


1.57 


1.60 


24-25 


11th.... 


10.25 


10.45 


1.56 


1.60 


25-26 


12th.... 


10.13 


10.11 


1.47 


1.51 


26-27 


13th 


J 9.91 


10.00 


*1.45 


1 1.62 


27-28 


14th 


10.43 


10.25 


1.57 


1.61 


28-29 


15th.... 


8.46 


8.58 


1.43 


1.46 


29-30 


16th 


9.58 


9.47 


1.91 


1.97 


Apr. 30-May 1 


17th 


8.81 


8.77 


1.90 


1.93 


May 1-2 


18th 

19th.... 

21st 

24th 

25th 

26th 


8.27 
8.37 
7.93 
8.15 
7.81 
7.88 


8.45 
8.11 
7.90 
8.24 
7.91 
8.00 


1.80 
1.76 

1.52 
1.51 
1.42 


1.80 
1.81 

1.55 
1.52 
1.43 


2-3 


4-5 


7-8 


8-9 


9-10 


10-11 


27th.... 


8.07 


7.99 


1.36 


1.39 


11-12 


28th.... 


7.62 


7.46 


1.28 


1.29 


12-13 


29th.... 


7.54 


7.64 


1.32 


1.32 


13-14 


30th 


7.83 


7.60 


1.32 


1.32 


14-15 


31st 


6.94 


6.83 


1.25 


1.22 


16-16 




3.72 
3.81 


3.64 
3.98 


.52 
.36 


.53 
.36 


16-17 









1 For total amounts of nitrogen and ammonia-N on this day, see table 29. 



254 A STUDY OF PROLONGED FASTING. 

known substances, such as ammonium-sulphate, ammonium ferrous 
sulphate, urea, and uric acid. We were thus assured of the accuracy 
of the methods. Since the Folin microchemical methods played such 
an important role in these analyses, particularly in an economical 
distribution of the available urine, it seems desirable to publish the 
results of the control tests. Accordingly, in table 28, the values 
obtained for the total nitrogen by the Kjeldahl method are compared 
with those secured by the microchemical methods; the values for the 
ammonia-nitrogen obtained with the new and old Folin methods are 
also compared. As will be seen, the results of such comparison are 
most satisfactory. We wish again to emphasize the great value of 
these methods, particularly when there is urgent necessity for the use of 
small samples. 

THE PARTITION OF THE NITROGEN EXCRETION. 

While the total nitrogen excretion in the urine of a fasting man has 
a general interest, more particularly in the apportionment of the total 
energy requirement and the energy output among the various factors, 
protein, carbohydrate, and fat, a clear understanding of the nature of 
the disintegration of the nitrogenous material is obtained only when a 
partition of the nitrogen excretion is made according to the analytical 
scheme of Folin. Fortunately, with all of the samples of urine col- 
lected for the 31 days of the fast we were able to secure a complete 
partition of the nitrogen, with the single exception of the determination 
of the total purines; we were, however, able to determine the uric 
acid-nitrogen. This partition included the determination of the total 
nitrogen, and the nitrogen from urea, ammonia, uric acid, creatinine 
preformed, and total creatinine. The nitrogen undetermined is given 
as "rest nitrogen." Furthermore, since Folin has shown the great 
significance of the proportionate distribution of the nitrogen derived 
from these various sources, we have computed the percentage of the 
total nitrogen in these nitrogenous constituents of the urine. The 
values for each day, expressed in grams and in percentages of total 
nitrogen, are given in table 29. 

Urea. 

With the microchemical method of Folin, the urea-nitrogen in the 
fasting urines could be determined with great accuracy. The deter- 
minations were accordingly made by Mr. Higgins for the 31 days of the 
fasting period and for the 3 days following the fast. The results are 
given in table 29, together with the percentage of urea-nitrogen in 
terms of total nitrogen. 

In general the course of the excretion of the urea-nitrogen follows 
quite closely that of the total nitrogen. The regular increase shown 
in the first 3 days for the total nitrogen is also apparent here; the 



URINE. 



; h v 






255 



smallest amount of urea-nitrogen (4.84 grams) is likewise found on the 
last day of the fast, but unlike the total nitrogen is considerably smaller 
than the excretion for the first day of the fast. The percentage of urea- 
nitrogen shown for the first few days, i. e., approximately 80 per cent, 
is distinctly lower than that found with normal urine, Folin 1 giving as 



Table 29. — Partition of nitrogen excreted in urine in experiment with L. 


Date. 


Day of 
fast. 


Excretion of nitrogen. 


Total 
nitrogen. 


Urea-N. 


Ammo- 
nia-N. 


Uric 
acid-N. 


Creati- 
nine-N 
(pre- 
formed). 


Total 
creati- 
nine-N. 


Rest-N. 


1912. 




gm. 


gm. 


gm. 


gm. 


gm. 


gm. 


gm. 


Apr. 11-12. . . 




15.92 

14.48 

11.54 

7.10 


5.68 


0.67 
.65 
.59 
.41 


0.112 


6!51 


0.48 


0.42 


12-13 . . . 




13-14 . . . 




14-15 . . . 


1st 


15-16... 


2d 


8.40 


6.69 


.60 


.049 


.46 


.46 


.60 


16-17... 


3d 


11.34 


9.11 


.95 


.042 


.46 


.55 


.69 


17-18... 


4th.... 


11.87 


9.03 


1.40 


.044 


.42 


.54 


.86 


18-19... 


6th.... 


10.41 


7.58 


1.62 


.059 


.41 


.51 


.64 


19-20. .. 


6th.... 


10.18 


7.36 


1.68 


.097 


.39 


.52 


.52 


20-21 . . . 


7th.... 


9.79 


7.02 


1.54 


.112 


.38 


.49 


.63 


21-22 . . . 


8th.... 


10.27 


7.45 


1.65 


.108 


.38 


.50 


.56 


22-23 . . . 


9th.... 


10.74 


7.83 


1.69 


.099 


.37 


.50 


.62 


23-24... 


10th.... 


10.05 


7.44 


1.59 


.118 


.37 


.49 


.41 


24-25. .. 


11th.... 


10.25 


7.66 


1.58 


.116 


.37 


.49 


.40 


25-26.. . 


12th 


10.13 


7.43 


1.49 


.154 


.37 


.49 


.57 


26-27... 


13th 


10.35 


7.69 


1.55 


.093 


.35 


.48 


.54 


27-28... 


14th.... 


10.43 


7.69 


1.59 


.125 


.33 


.44 


.59 


28-29 . . . 


15th.... 


8.46 


6.18 


1.45 


.071 


.30 


.38 


.38 


29-30... 


16th 


9.58 


6.71 


1.94 


.099 


.32 


.42 


.41 


Apr. 30-May 1 


17th.... 


8.81 


5.95 


1.92 


.100 


.31 


.40 


.44 


May 1- 2 . . . 


18th 


8.27 


5.70 


1.80 


.122 


.34 


.41 


.24 


2- 3 . . . 


19th.... 


8.37 


5.58 


1.79 


.130 


.30 


.38 


.49 


3-4... 


20th.... 


7.69 


5.36 


1.58 


.115 


.31 


.38 


.26 


4-6... 


21st 


7.93 


5.54 


1.57 


.112 


.31 


.38 


.33 


5- 6. .. 


22d. , . 


7.75 


5.60 


1.51 


.110 


.31 


.36 


.17 


6-7... 


23d, 


7.31 


5.01 


1.49 


.097 


.34 


.36 


.35 


7-8... 


24th.... 


8.15 


5.92 


1.54 


.114 


.30 


.34 


.24 


8-9... 


25th.... 


7.81 


5.43 


1.52 


.098 


.28 


.35 


.41 


9-10... 


26th 


7.88 


5.62 


1.43 


.063 


.29 


.36 


.41 


10-11. . . 


27th 


8.07 


5.90 


1.38 


.089 


.29 


.35 


.35 


11-12. . . 


28th 


7.62 


5.46 


1.29 


.095 


.28 


.34 


.44 


12-13.. . 


29th 


7.54 


5.55 


1.32 


.101 


.29 


.35 


.22 


13-14... 


30th.... 


7.83 


5.53 


1.32 


.106 


.29 


.33 


.54 


14-15... 


31st 


6.94 


4.84 


1.24 


.122 


.30 


.32 


.42 


15-16. . . 




4.83 
3.81 
2.75 


3.21 
2.69 
1.54 


.69 
.36 
.35 


.140 
.144 
.111 


.35 
.34 
.35 


.37 
.34 
.33 


.42 
.28 
.42 


16-17. . . 




1 17-18 . . . 









x The amounts for this day were determined in the urine for about 22 hours. 

an average for his subjects 87.5 per cent of the total nitrogen in the form 
of urea. The percentage of the urea-nitrogen then falls rapidly until 
on the fifth day it is but 72.82 per cent. The lowest percentage found 

^olin, Am. Journ. Physiol., 1905, 13, p. 62. 









256 A STUDY OF PROLONGED FASTING. 

Table 29. — Partition of nitrogen excreted in urine in experiment tvith L. — Continued. 



Date. 


Day of 


Proportion of total nitrogen in — 














fast. 


Urea. 


Ammo- 
nia. 


Uric 
acid. 


Creatinine 

(pre- 
formed). 


Total 
creatinine. 


Rest-N. 


1912. 




p. ct. 


p. ct. 


p. ct. 


p. ct. 


p. ct. 


p. ct. 


Apr. 11-12 






4.21 










12-13 




80.00 


4.49 
6.11 
5.77 


1.58 


7.18 


6.76 


5.89 


13-14 




14-16 


1st 


16-16 


2d 


79.64 


7.14 


.58 


5.48 


5.48 


7.16 


16-17 


3d 


80.33 


8.38 


.37 


4.06 


4.85 


6.07 


17-18 


4th.... 


76.07 


11.79 


.37 


3.54 


4.55 


7.22 


18-19 


6th ... . 


72.82 


15.66 


.67 


3.94 


4.90 


6.15 


19-20 


6th 


72.30 


16.50 


.95 


3.83 


5.11 


5.14 


20-21 


7th.... 


71.71 


15.73 


1.14 


3.88 


5.01 


6.41 


21-22 


8th.... 


72.54 


16.07 


1.05 


3.70 


4.87 


5.47 


22-23 


9th 


72.90 


15.73 


.92 


3.44 


4.66 


5.79 


23-24 


10th 


74.03 


15.82 


1.17 


3.68 


4.88 


4.10 


24-25 


11th.... 


74.73 


15.41 


1.13 


3.61 


4.78 


3.95 


25-26 


12th 


73.35 


14.71 


1.52 


3.65 


4.84 


5.58 


26-27 


13th.... 


74.30 


14.98 


.90 


3.38 


4.64 


5.18 


27-28 


14th 


73.73 


15.24 


1.20 


3.16 


4.22 


5.61 


28-29 


15th 


73.05 


17.14 


.84 


3.55 


4.49 


4.48 


29-30 


16th 


70.04 


20.25 


1.03 


3.34 


4.38 


4.30 


Apr. 30-May 1 


17th.... 


67.54 


21.79 


1.14 


3.52 


4.54 


4.99 


May 1-2 


18th.... 


68.92 


21.77 


1.48 


4.11 


4.96 


2.87 


2-3 


19th 


66.67 


21.39 


1.55 


3.58 


4.54 


5.85 


3-4 


20th 


69.70 


20.55 


1.50 


4.03 


4.94 


3.31 


4- 5 


21st 


69.86 


19.80 


1.41 


3.91 


4.79 


4.14 


5-6 


22d 


72.26 


19.49 


1.42 


4.00 


4.65 


2.18 


6-7 


23d 


68.54 


20.38 


1.33 


4.65 


4.92 


4.83 


7-8 


24th.... 


72.64 


18.90 


1.40 


3.68 


4.17 


2.89 


8- 9 


25th 


69.52 


19.46 


1.25 


3.69 


4.48 


5.29 


9-10 


26th 


71.32 


18.15 


.80 


3.68 


4.57 


5.16 


10-11 


27th 


73.12 


17.10 


1.10 


3.59 


4.34 


4.34 


11-12 


28th 


71.66 


16.93 


1.25 


3.67 


4.46 


5.70 


12-13 


29th 


73.61 


17.51 


1.34 


3.85 


4.64 


2.90 


13-14 


30th 


70.63 


16.86 


1.35 


3.70 


4.21 


6.95 


14-15 


31st 


69.74 


17.87 


1.76 


4.32 


4.61 


6.02 


15-16 




66.46 
70.60 
56.00 


14.28 

9.45 

12.73 


2.90 
3.78 
4.04 


7.25 

8.92 

12.73 


7.66 

8.92 

12.00 


8.70 

7.25 

15.23 


16-17 




U7-18 









*The amounts for this day were determined in the urine for about 22 hours. 

during the last 26 days was 66.67 per cent on the nineteenth day and 
the highest on the eleventh day of 74.73 per cent. The average value 
for these days was 71.5 per cent, with a distinct tendency towards 
constancy. Upon the resumption of food, there was at first no marked 
disturbance in this ratio, but on the third day the percentage of urea- 
nitrogen decreased to the low value of 56 per cent, this value being 
found at the time that only 2.75 grams of nitrogen were excreted in 
the urine. These low values in the percentage of urea-nitrogen are 
perfectly comparable with those found by Folin on subjects subsisting 



URINE. 257 

on a starch-cream diet, with an excretion of nitrogen corresponding 
to not far from 4 to 5 grams per day. 

These figures are also substantiated by the observations of Cathcart. 
While he found in the first 3 days of Beauty's fast that the urea-nitro- 
gen averaged not far from 87 per cent instead of the 80 per cent found 
with our subject L., and that the values also averaged somewhat higher 
for the remainder of the fast, nevertheless the percentage fell as low as 
71 per cent on the eighth day of the fast. On the food days following 
the fast, the urea-nitrogen fell to 61.97 per cent on the day when the 
minimum nitrogen excretion was observed. 

E. and 0. Freund found in their observations on Succi that the urea- 
nitrogen was 82 per cent or more of the total nitrogen excretion for 
the first two weeks of the fast. There was then a rapid fall in the 
percentage until but 56 and 58 per cent of urea-nitrogen were found 
on the last 2 days. Brugsch's observations on Succi in Hamburg show 
that for the last 8 days of the 30-day fast, the urea-nitrogen was not 
far from 60 per cent of the total nitrogen. 

Van Hoogenhuyze and Verploegh, in their observations on Tosca, 
note most irregular proportions of urea-nitrogen. The percentages 
of urea-nitrogen computed by us from their data are as follows, the 
dayof the fast being given in parentheses: (1)68.84; (2)79.11; (3)93.29; 
(4) 84.90; (5) 66.07; (6) 59.00?; (7)81.02; (8)84.42; (9)88.43; (10)87.65; 
(11) 85.84; (12) 86.80; (13) 86.84; (14) 88.73. On the sixth day the 
low value of 59 per cent is from data questioned by the authors. In 
the light of other fasting studies, there is no obvious explanation 
for the unusually high average value, especially for the last 8 days of 
the fast. 

From all the evidence it can be seen that in general during fasting 
the urea output approximately parallels that of the total nitrogen, 
there being a decided increase on the first few days of fasting, followed 
by a decrease. In practically every instance when there is a fluctua- 
tion in the total nitrogen output, this is paralleled by the urea-nitrogen. 
It would thus appear that the determining factor in the fluctuations 
of the total nitrogen is probably the proportion of urea-nitrogen, and 
not the gross alterations in the other factors. The variation in the 
percentage distribution can, however, be intelligently treated only after 
a consideration of the changes in the output of ammonia-nitrogen. 

Ammonia. 

The ammonia-nitrogen, on account of its great significance in con- 
sidering the products of defective fat katabolism, was determined by 
both the old and the new Folin methods. The results of these two 
series of determinations are given in table 28, page 253. An average of 
these two series of values is also given in table 29. 

Normal urine always contains a relatively small amount of ammonia, 
and the amounts found for L. on the 3 days prior to the fast were 



258 A STUDY OF PROLONGED FASTING. 

approximately those which would be noted for normal individuals 
subsisting on a diet containing about 15 grams of nitrogen. The 
ammonia-nitrogen formed not far from 4.5 per cent of the total nitrogen 
excretion per day. At the beginning of the fasting period the amount 
of ammonia-nitrogen excreted fell somewhat, and not until the third 
day do we find values exceeding those obtained before the fast. On 
the fourth day it rose quite sharply to 1.40 grams and then continued 
to rise steadily, with slight fluctuations, until the maximum value of 
1.94 grams was reached on the sixteenth day. Thereafter it slowly 
and quite regularly decreased until the end of the experiment, the 
excretion of ammonia-nitrogen on the last day being 1.24 grams. This 
gradual increase and decrease was exactly that observed by Cathcart, 
although the maximum value with his subject was observed on the 
eighth day, while with L. it did not appear until the sixteenth day. 
Brugsch found the excretion of ammonia-nitrogen quite regular, rang- 
ing between 1.26 grams and 1.72 grams in the last 8 days of Succi's 
fast in Hamburg. 

Since the amount of ammonia-nitrogen would normally be expected 
to fluctuate somewhat with the fluctuations in the total nitrogen 
excretion, the percentage of ammonia-nitrogen in the total nitrogen 
must be considered. On this basis the minimum percentage (5.77 per 
cent) is found on the first day of the fast, with the maximum (21.79 
per cent) on the seventeenth day. Even on the percentage basis the 
ammonia-nitrogen tends to increase until the middle of the fast and 
then slowly to decrease, although towards the end it was still 10 per 
cent or more larger than it was prior to the fasting period. 

It is clear, therefore, that there was some factor which stimulated 
the excretion of ammonia-nitrogen. From previous experience with 
fasting subjects, it is obvious that this increase in the excretion was 
due to the organic acids, chiefly the /3-oxybutyric acid resulting from 
defective fat katabolism. The large excretion of ammonia therefore 
undoubtedly corresponds to an increasing acidosis, the production of 
the ammonia being a protective action on the part of the body to over- 
come the effect of the acids. The amounts of ammonia-nitrogen found 
for L. during fasting were measurably greater than those observed by 
Cathcart for his subject and the percentage of the total nitrogen was 
also considerably greater. These values would therefore imply that L. 
had a somewhat greater acidosis than had Beaute" in Cathcart's research. 

The results of the observations of E. and O. Freund on Succi are 
diametrically opposed to the values found by Cathcart on Beauts, 
Brugsch on Succi, and by us with L., for both the absolute amount and 
the percentage of the total nitrogen found, and one is inclined to ques- 
tion somewhat the technique of the Freunds. While the observations of 
Bonniger and Mohr 1 on the fasting woman Schenk are complicated by the 

1 Bonniger and Mohr, Zeitschr. f. exp. Path. u. Therapie, 1906, 3, p. 675. 



URINE. 259 

introduction of amino-acids, their results shoul d also be cited. The abso- 
lute amounts of ammonia-nitrogen excreted varied from 0.40 gram on 
the second fasting day to 1 .84 gram on the seventh day of fasting. The 
percentage of ammonia-nitrogen in the total nitrogen varied from 5.28 
per cent on the second day to the high value of 30.57 per cent on the 
sixteenth day of the fast. In general, these values are not unlike those 
observed by us on our subject L. 

If we compare the values obtained with L. for the urea-nitrogen and 
the ammonia-nitrogen, it is evident that the ammonia-nitrogen was 
formed at the expense of urea-nitrogen, for whenever the values for 
the urea-nitrogen decrease, those for the ammonia-nitrogen increase, 
and vice versa. A fact of particular interest is the rapid return to a 
small excretion of ammonia-nitrogen with the taking of food by the 
subject. Even on the first day with food, the ammonia-nitrogen fell 
to 0.69 gram. On subsequent days the absolute values decreased to 
0.36 gram and 0.35 gram respectively, though still forming 10 to 12 per 
cent of the total nitrogen. It is obvious, therefore, that, with the inges- 
tion of a large amount of carbohydrate and a decrease in the acidosis, 
there was no necessity for an excessive ammonia-nitrogen excretion. 

Uric Acid. 

In the earlier fasting experiments at Wesleyan University, material 
was available for only a few determinations of the uric acid, which were 
made possible through the courtesy of Professor Lafayette B. Mendel, 
of Yale University. In this experiment with L. the determinations 
of uric acid were personally made by Professor Otto Folin, by his new 
colorimetric method, 1 small specimens of the 24-hour urine being sent 
to the Harvard Medical School daily for analysis. The results of his 
determinations, expressed as uric-acid nitrogen, are given in table 29. 

No determinations of the uric-acid nitrogen were made prior to the 
fasting period. During the fast the minimum amount of 0.042 gram 
Was obtained on the third day and the maximum of 0.154 gram on the 
twelfth day, the amounts varying considerably throughout the entire 
fasting period. 

With the origin of uric acid still the subject of considerable critical 
debate, particularly between Siven 2 on the one hand and Mares 3 and 
his colleagues on the other, it can be seen that although the general 
tendency is for physiologists to uphold the views of Mares, 4 it is difficult 
to interpret the values for the uric-acid excretion obtained in this fast 
on any of the present hypotheses as to its origin. That it is a deriva- 

^olin and Denis, Journ. Biol. Chem., 1913, 14, p. 95. 
2 Siven, Archiv f. d. ges. Physiol., 1912, 146, p. 499. 

3 Mare§, Archiv f. d. ges. Physiol., 1912-1913, 149, p. 275. See also Smetdnka, Archiv f. d. 
ges. Physiol., 1911, 138, p. 217; ibid., 1912-1913, 149, p. 287. 
4 Mares, Archiv f. d. ges. Physiol., 1910, 134, pp. 59-102 



260 A STUDY OF PROLONGED FASTING. 

tion of nuclein katabolism is undoubtedly true. Beyond this univer- 
sally accepted fact, the evidence as to the influence of intestinal, 
glandular, kidney, and muscular activity is much debated. 

In previous fasting experiments it has been shown that the glandular 
activity, at least so far as the digestive organs are concerned, is at a 
minimum, 1 and hence we should expect to find that if any considerable 
proportion of the uric acid were obtained as the result of glandular 
activity, there would be a minimum uric-acid nitrogen excretion during 
the fast. Aside from the admittedly low values on the second to the 
fifth days of the fasting experiment with L., this was not the case, for 
there are many days during the fast on which there was fully as much 
uric-acid nitrogen, if not indeed more, than one would expect to find 
with a normal individual subsisting on a purine-free diet. 2 

After L. had taken food, there was a slight increase in the total 
uric-acid nitrogen excretion from 0.122 gram, which was obtained on 
the last day of the fast, to 0.140 gram on the first day with food. Ordi- 
narily this would not be considered as a substantial increase in the 
total uric-acid nitrogen excretion, and yet in recent controversial 
papers, Mares has pointed out that a 10 per cent increase is not to be 
ignored and that it does represent an actual increase. While accord- 
ing to Mares, the ingestion of food which stimulates the digestive 
glands to their greatest activity is most productive of an increase in 
the excretion of uric acid, nevertheless, Smetanka has found that such 
an excretion took place even with honey. Inasmuch as the food taken 
by L. was in large part carbohydrate, a portion of the diet being 
honey, it is not impossible to believe that there may have been 
a positive increase in the uric-acid nitrogen as a result of the ingestion 
of this food. This is the more credible if we give heed to Smetanka's 
criticism that 24-hour periods are not best suited to the study of this 
problem, as the influence of the activity of the digestive glands dis- 
appears rapidly after the ingestion of food. 

During the fasting period, the percentage of the total nitrogen in 
the form of uric-acid nitrogen likewise underwent considerable fluctua- 
tion, approximately paralleling the absolute amount of this constituent. 
The percentage value shows a distinctly increased uric-acid nitrogen 
excretion in the post-fasting period, rising to 4.04 per cent, and while 
the evidence is by no means complete, this may well be interpreted 
as an index of an increased glandular (digestive) activity. 

Cathcart also determined the uric-acid nitrogen excretion for his 
subject, using the method of Hopkins-Folin. During the first half 
of the fast he found a distinctly lower value for the uric-acid nitrogen 
than in the preceding food period, but there was a continual tendency 
for this form of nitrogen to increase as the fast progressed. With the 
taking of food there was an increase in the uric-acid nitrogen — at 

'Luciani, Das Hungem, Hamburg and Leipsic, 1890, p. 44. 
2 Folin, Am. Journ. Physiol., 1905, 13, p. 62. 



URINE. 261 

least on the first day — followed by 4 days with the same excretion as 
in the last period of the fast. Thus, in Beauty's fast there was a 
definite tendency towards regularity 1 in the excretion of the uric-acid 
nitrogen, inasmuch as there was from the beginning to the end a 
regular, though slight, increase — a regularity that was by no means 
paralleled by the observations on L. 

While the values found for L. were decidedly variable, yet in several 
points they strikingly confirm the earlier observations. Thus the 
marked fall on the second fasting day, followed by a low excretion for 
several days and a subsequent rise, with a higher value on the first 
food day than that shown on the last fasting day, are all in agreement 
with the observations of Cathcart with Beaute, and of Van Hoogen- 
huyze and Verploegh in their 14-day experiment with Tosca. 

The rapid fall in the uric-acid nitrogen excretion on the second day 
has also been observed in shorter fasts by Schreiber and Waldvogel, 2 
by Hirschstein, 3 and by Feldmann. 4 Scaffidi, 5 in experimenting on 
the purine metabolism in fasting, concluded that with those animals 
with which there was a formation of oxidative uric acid, the uric-acid 
nitrogen decreased during starvation and there was no regularity in 
the relation of total uric-acid nitrogen to the total nitrogen. 

This decrease in the uric-acid excretion of L. was coincident with 
that period of the fast when the supply of glycogen was rapidly being 
depleted and the subsequent increase followed sharply the incidence 
of the protein-fat katabolism characteristic of the remainder of the 
fast. The cessation of the glandular activity of digestion on the first 
few days may explain the fall in the uric-acid nitrogen, but the sub- 
sequent increase can only be explained by an increase in the katabolism 
of the active protoplasmic tissue. The fact that this excretion does not 
remain constant, or at least does not regularly increase or decrease, 
is indeed difficult to explain, since there are no obvious reasons for 
assuming that the changes in the amount of the excretion of uric-acid 
nitrogen are due to corresponding changes in the rate of destruction 
of the active protoplasmic tissue. Indeed, if we observe the total 
creatinine excretion, the regular decrease of that urinary constituent 
would imply a proportional regularity in the rate of destruction of the 
active protoplasmic tissue. It is evident that the values obtained for 
the excretion of the uric-acid nitrogen in this fasting experiment offer 
no proof of the validity of any of the present-day conceptions as to 
the origin of endogenous uric-acid excretion. 

Regularity in the uric-acid excretion as the fast progressed was also noted for the greater part 
of the experiment with Tosca by Van Hoogenhuyze and Verploegh (loc. cit.) and with Succi in 
Vienna by E. and O. Freund (Wiener klin. Rundsch., 1901, 15, p. 69). Brugsch (Zeitschr, f. 
exp^Path. u. Therapie, 1905, 1, p. 419) found very constant values for the purine-nitrogen during 
the last eight days of Succi's Hamburg fast. 

2 Schreiber and Waldvogel, Arch. f. exp. Path. u. Pharm., 1899, 42, p. 69. 

'Hirschstein, Arch. f. exp. Path. u. Pharm., 1907, 57, p. 229. 

4 Feldmann, cited by Siven, Archiv. f. d. ges. Physiol., 1912, 146, p. 499. 

'Scaffidi, Biochem. Zeitschr., 1911, 33, p. 153. 



262 A STUDY OF PROLONGED FASTING. 

Creatinine. 

As a result of Folin's fundamental observations on the partition of 
the nitrogen of normal urines, special stress was laid upon the deter- 
mination of the creatinine existing in the urine of this fasting man, 
Folin having emphasized the fact that the total creatinine may be 
looked upon as an index of the total tissue metabolism. 1 Immediately 
after Folin's papers had appeared, a large number of researches on the 
creatinine in urine were reported, and observations were simultane- 
ously made in several laboratories which implied the presence of 
creatine in urine under certain conditions, particularly in pathological 
cases and during fasting. 

Folin's analytical scheme enabled the direct determination of crea- 
tine, the creatinine being first determined, and subsequently any creatine 
in the urine was converted to creatinine by heating with acid for 3 
hours. These results were reported by him as creatinine preformed 
and total creatinine, the difference in the two values being estimated 
to be a measure of the creatine expressed in terms of creatinine. 

There is little of positive value to be said regarding the determinations 
of creatine or creatinine in fasting experiments prior to the introduction 
of the Folin method, hence Baldi's 2 observations as well as those of the 
Freunds 3 on Succi can have but little quantitative interest. One of the 
first studies made after the Folin method was put forth is that of Van 
Hoogenhuyze and Verploegh on Tosca. 4 The considerable decrease in 
the creatinine obtained by them at the beginning of the fast, followed 
by an increase after food, shows (in part at least) that only preformed 
creatinine was determined, and hence their values correspond more 
nearly to the values reported for our subject as preformed creatinine. 

In the experiments at Wesleyan University, the total creatinine 
was invariably found to be somewhat higher in the fasting periods 
than the preformed creatinine, the difference disappearing when food 
was taken, and thus the conclusion was drawn that creatine was to be 
found regularly in the urine of fasting subjects. This observation was 
simultaneously made and published by Cathcart. 5 Laying special 
emphasis upon the difference between preformed and total creatinine, 
Cathcart discussed at considerable length the appearance of creatine 
in the urine. Although our interpretation of the presence of creatine 
in the urine differs considerably from that of Cathcart, the fact remains 
that since that time considerable research has been carried out on both 
creatinine and creatine excretion, and hence an observation of the 
amount excreted by our fasting subject was particularly desirable. 

^olin, Am. Journ. Physiol., 1905, 13, pp. 66 and 117. 

2 Baldi, Sperimentale, March 1889; Centrlb. f. klin. Med., 1889, 10, p. 651. 

3 E. and O. Freund, Wiener klin. Rundsch., 190 1, 15, pp. 69 and 91. 

4 Van Hoogenhuyze and Verploegh, Zeitschr. f. physiol. Chemie, 1905, 46, p. 415. 

'Cathcart, Biochem. Zeitschr., 1907, 6, p. 109. 



URINE. 263 

The only other fasts on human subjects in which the Folin methods 
have been used are the two 7-day fasts reported by Howe, Mattill, 
and Hawk. 1 Here, as in our experiments, preformed creatinine and 
total creatinine were determined without prior treatment to remove 
aceto-acetic acid. The large amount of creatine-nitrogen found by 
them on the first fasting day with their subject E., namely, 0.269 gram, 
is wholly inexplicable, being larger than any observation with which we 
are familiar. After the first 3 days there was a regular decrease in 
the creatine-nitrogen excretion, until on the seventh day it was prac- 
tically nothing. With subject H. they found an increasing creatine- 
nitrogen excretion for the first 4 days, followed by a decrease. It 
is thus seen that, as the result of determinations by the Folin method, 
several observers have noted that the total creatinine is greater than 
the preformed creatinine. 

In interpreting the increase in the creatinine excretion in the fasting 
experiments at Wesleyan University as being due to the presence of 
creatine, attention was especially directed 2 to the possible influence 
upon the Jaffe" reaction of substances other than creatine and creati- 
nine. Immediately after this publication had appeared, and in sub- 
sequent visits to foreign laboratories where work upon creatine and 
creatinine was being carried out, I was assured that no substance 
existing in either fasting or diabetic urine appreciably affected the 
reaction and that this interpretation of the difference between pre- 
formed creatinine and total creatinine was undoubtedly justifiable. 
Accordingly, in presenting the results obtained with L., the assumption 
has been made that a difference between preformed creatinine and 
total creatinine, as ordinarily determined, could be taken as an index 
of the presence of creatine. 

The creatinine determinations in these fasting urines were made by 
Miss Alice Johnson, under the immediate supervision of Dr. A. W. 
Peters, numerous control tests being made with samples of pure creati- 
nine kindly furnished by Professor Otto Folin. The preformed creati- 
nine was determined according to the usual Folin method and in no 
instance was abnormal fading or alteration in color noted. Further- 
more, in the light of recent investigations, it is important to note that 
no special treatment was given for the removal of acetone bodies which 
would possibly have affected the color. After the determination of the 
preformed creatinine, the total creatinine was determined by heating 
another specimen of the urine with hydrochloric acid for 3 hours on 
an electric plate to convert the creatine to creatinine. The readings 
were carefully controlled and the illumination of the colorimeter was 
given special attention. The results as found by these two processes 
are expressed as nitrogen of creatinine preformed and total creatinine 

'Howe, Mattill, and Hawk, Journ. Am. Chem. Soc, 1911, 33, p. 568. 
2 Benedict, Carnegie Inst. Wash. Pub. 77, 1907, p. 395. 



264 A STUDY OF PROLONGED FASTING. 

and are given in table 29. As with the earlier fasting experiments, 
there was a difference between the creatinine preformed and the total 
creatinine as thus measured. 

At the time these analyses were made, we were firmly of the opinion 
that the results obtained by such analysis would represent quanti- 
tatively the amount of creatine in urine, expressed as creatinine. 
Since these investigations were carried out, two papers have appeared 
which lead us to question the quantitative relations exhibited in the 
values here given. Greenwald, 1 in his work on diabetic urines, con- 
siders the effect of the acetone bodies upon the Jaffe reaction and con- 
cludes that urine containing aceto-acetic acid and acetone will give 
correct results for creatinine by the Folin method only after the 
removal of these substances. More recently Graham and Poulton 2 
have studied the influence of aceto-acetic acid on the estimation of 
creatinine and have called into question the interpretation of the dif- 
ference between preformed and total creatinine as reported in fasting 
experiments and in experiments with carbohydrate starvation. They 
conclude that " acetone and /3-oxybutyric acid, if present in amounts 
comparable to those which usually occur in urine, produce practically no 
error in the estimation of creatinine." They furthermore state that 
aceto-acetic acid causes an error in the preformed creatinine determina- 
tion, but does not affect the determination of total creatinine. On 
the basis of their experiments, they conclude that the difference 
between creatinine preformed and total creatinine does not represent 
creatine, since in their experiments with carbohydrate-free diets, they 
found no excretion of creatine. 

It is obvious, therefore, that these two researches throw considerable 
doubt upon the value of the figures reported under the head of the 
preformed creatinine, but affect in no wise the values for the total 
creatinine. We may therefore without further reservation proceed to 
a consideration of the total creatinine-nitrogen as reported in table 29. 

One of the most striking features of these results is their great regu- 
larity. After the first 2 days there was an almost continuous fall in 
the total creatinine-nitrogen from the maximum on the third day of 
0.55 to a minimum on the last day of the fast of 0.32. The minor 
fluctuations from this regular fall are so few as to be negligible. 

In the fasting experiments with S. A. B. at Wesleyan University, 
the almost absolute constancy in the total creatinine-nitrogen was a 
matter of special comment and in the report of the results it was 
pointed out that it was probably more than a mere coincidence that 
the sum of the creatine-nitrogen and the preformed creatinine-nitrogen 
remained constant each day as the fast progressed. It will be seen 
that for the first 7 days of the fast with L., while the results were not 

Greenwald, Journ. Biol. Chem., 1913, 14, p. 87. 

Graham and Poulton, Proc. Royal Soc, ser. B., 1914, 87, p. 205. 



URINE. 265 

constant, nevertheless the amount excreted did not vary greatly, 
although the variation was wholly in the line of a decrease toward the 
end. It was only after the thirteenth day that any considerable de- 
crease in the total creatinine-nitrogen appeared; from that time it 
remained at a lower level, gradually decreasing until the end of the 
fast, the lowest value being found on the last day of the fast. 

Of particular interest is the striking regularity in the percentage of the 
total nitrogen excreted in the form of total creatinine. Omitting the 
first 2 days, the percentage ranged from 5.11 per cent on the sixth day 
to 4.17 per cent on the twenty-fourth day. 

On the first 2 days of the fast, the values for preformed creatinine- 
nitrogen and for total creatinine-nitrogen are essentially the same, 
and it is only with the third day that we begin to find the measurable 
differences which have been ascribed to creatine. In an attempt to 
explain the presence of creatine in the fasting urine, two hypotheses 
have been presented: one, that as a result of inanition the body loses 
its power of converting creatine to creatinine before excreting it, and 
the other that creatine was representative of the flesh katabolized, as 
in the disappearance of the body material a certain amount of creatine 
normally existing in the flesh was liberated and excreted in the urine. 
Against the validity of this latter assumption was the obviously sig- 
nificant fact that the sum of the creatinine-nitrogen and the creatine- 
nitrogen remained constant. While without doubt the observations 
of Greenwald and of Graham and Poulton affect the quantitative value 
of the difference between preformed creatinine and total creatinine, the 
accumulative evidence of the past eight years is such as to make it 
reasonably certain that creatine is excreted unchanged in fasting and 
pathological urines and our uncertainty lies, therefore, only in our 
knowledge as to the quantities thus excreted. 

In a recent paper Folin 1 has reiterated his belief regarding the 
interpretation of the creatinine output in the following terms: "The 
creatinine elimination becomes more clearly than ever the most clear- 
cut index or measure of the total normal tissue metabolism." In this 
paper Folin explains the appearance of creatine as an abnormal split- 
ting off of a cleavage product which is normally excreted as creatinine 
and disclaims the presence of isolated, uncombined creatine in flesh, 
admitting that in fasting and in various pathological conditions the 
normal breaking down into creatinine is accompanied more or less by 
an abnormal breaking down into creatine. It is clear, therefore, that 
according to all of Folin's recent interpretations the exact quantitative 
knowledge regarding the creatine in urine has no longer the significance 
formerly attributed to it, and hence the criticisms of Greenwald and of 
Graham and Poulton, while admittedly justifiable so far as technique 
is concerned, apply to a determination which bids fair to have but 

^olin and Denis, Journ. Biol. Chem., 1914, 17, p. 501. 



266 A STUDY OF PROLONGED FASTING. 

little physiological significance. The pathological significance must, 
however, be measurably increased in value as a result of their researches. 

Of fundamental importance is the possibility of using Folin's inter- 
pretation of the total creatinine excretion as an index of the metabolism 
of normal tissue. Throughout this entire monograph stress has been 
laid upon the importance of knowing, if possible, the active mass of 
protoplasmic tissue in the body, 1 and the question arises, "Have we in 
the total creatinine excretion an interpretable index of the changes 
in the active mass of protoplasmic tissue?" If, as is pointed out in 
the section on energy transformation, the active mass of protoplasmic 
tissue is the fundamental factor in the determination of the total 
energy requirement, we should normally expect to find that the total 
energy output decreased with the loss of creatinine, indicating a con- 
tinual decrease in the tissue metabolism. As a matter of fact, as the 
fast progresses, such a decrease in the total heat-production is clearly 
shown by various methods and yet, as a reference to the section on the 
pulse-rate will show, our whole conception of the relationship between 
creatinine and energy transformation is seriously affected by the 
increase in the pulse-rate found with this subject during the last week 
of the fast. The total energy transformation is clearly due to the 
active mass of protoplasmic tissue and the stimulus to cellular activity, 
but at no part of the fast were conditions so sharply differentiated that 
we may say with accuracy that the loss in the heat-production was 
directly comparable to the loss in the normal katabolized tissue, as 
indicated by the decrease in the total creatinine excretion. Similarly, 
we find no definite relationship between the creatinine excretion and 
the total basal heat production, utilizing the creatinine elimination as 
an index of the total mass of active protoplasmic tissue remaining in 
the body. It is, however, of great significance that the total creatinine 
excretion decreased regularly as the fast progressed, thus indicating 
not an approximation to depletion but a distinct tendency on the 
part of the body to a conservation of its active protoplasmic tissue. 

The decreasing differences between the preformed creatinine and 
the total creatinine observable toward the end of the fast are not 
easily explained upon the ground of the influence of acetone bodies 
in the urine, since from the determinations of the ammonia and the 
/3-oxybutyric acid we have no reason to believe that the acidosis was 
materially less in the last week than at any other time. These values 
would therefore indicate that a sufficient amount of acetone bodies 
was not present in these urines to affect materially the quantitative 
determinations of preformed creatinine. 

1 During the reading of this proof, we received the admirable article, "Basal Metabolism and 
Creatinine Elimination," by W. W. Palmer, J. H. Means, and J. L. Gamble (Journ. Biol. Chem., 
1914, 19, p. 239). 



URINE. 



267 



With the taking of food, there was a slight increase in the total 
creatinine-nitrogen . on the first day, and the difference between the 
nitrogen of the creatinine preformed and of the total creatinine dis- 
appears after this day. 



Table 30. — Total creatinine excreted in urine, per day and per kilogram 
of body-weight, in experiment with L. 


Date. 


Day of fast. 


Creatinine excreted. 


Per kilogram 
of body-weight 
per day. ,. 


Per day. 
/ 


1912. 
Apr. 14-15 


1st 


• 

gm. ~ 

1.29 
1.23 
1.47 
1.45 
1.37 
1.40 
1.31 
1.35 
1.35 
1.31 
1.31 
1.31 
1.28 
1.19 
1.03 
1.14 
1.07 
1.09 
1.03 
1.01 
1.01 

.96 

.98 

.92 

.94 

.96 

.95 

.91 

.94 

.89 

.86 
1.00 

.91 
1 .88 


mg. 
21.5 
20.8 
25.2 
25.3 
24.2 
24.9 
23.5 
24.4 
24.6 
24.1 
24.3 
24.4 
23.9 
22.3 
19.4 
21.7 
20.6 
21.1 
20.1 
19.8 
19.9 
19.1 
19.6 
18.5 
19.0 
19.5 
19.4 
18.7 
19.5 
18.6 
18.1 
21.2 
19.3 
18.5 


15-16 


2d 


16-17 


3d 


17-18 


4th 


18-19 


5th 


19-20 


6th 


20-21 


7th 


21-22 


8th 


22-23 


9th 


23-24 


10th 


24-25 


11th 


25-26 


12th 


26-27 


13th 


27-28 


14th 


28-29 


15th 


29-30 


16th 


Apr. 30-May 1 


17th 


May 1-2 


18th 


2- 3 


19th 


3-4 


20th 


4- 5 


21st 


6- 6 


22d 


6-7 


23d 


7- 8 


24th 


8- 9 


25th 


9-10 


26th 


10-11 


27th 


11-12 


28th 


12-13 


29th 


13-14 


30th 


14-15 


31st 


15-16 




16-17 




17-18 









determined in urine for about 22 hours. 

As an indication of the relationship between the metabolism of 
active tissue and the body-weight — a relationship that must at best 
be somewhat approximate, depending in large part upon the composi- 
tion of the body — it has been customary to obtain a so-called creatinine 
coefficient by dividing the total creatinine excretion each day by the 
body-weight. This coefficient is normally found to be not far from 



268 A STUDY OF PROLONGED FASTING. 

20 to 30 milligrams. We have computed the creatinine coefficient for 
our subject L. for each day of the 31-day fast and give the values in 
table 30. After the first 2 days of fasting, the coefficient remains 
practically constant until the fourteenth day; it then shows a tendency 
to fall for a week, and the last 10 days it remains nearly constant at 
19 milligrams. 

Rest Nitrogen. 

When the urea-nitrogen, the ammonia-nitrogen, the uric-acid nitro- 
gen, and the total creatinine-nitrogen are combined, we find that the 
total amount is somewhat less than the total nitrogen found by th 
Kjeldahl method. This remainder, or the so-called "rest-nitrogen," 
amounts usually to not far from 0.5 gram of nitrogen in the observa- 
tions with our subject L. The values for this undetermined nitrogen 
are recorded in table 29, in which it is seen that the largest amount 
(0.86 gram) was found on the fourth day and the smallest (0.17 gram) 
on the twenty-second day. Since these values include all the errors in 
the analyses, the amounts thus recorded are not unexpected. Usually 
they form about 5 per cent of the total nitrogen and in these records 
vary from 7.22 per cent on the fourth day of the fast to 2.18 per cent 
on the twenty-second day. In the food period following the fast, we 
find a large increase on the percentage basis in this rest nitrogen, 
which reaches 15.23 per cent of the total nitrogen excretion. It thus 
appears that the chief factor affecting the nitrogen excretion on the last 
day was the urea, for the absolute amounts of ammonia-nitrogen, uric- 
acid nitrogen, creatinine-nitrogen, and rest-nitrogen remained essen- 
tially constant on all three days of food. 

ACID RADICLES. 

Fasting urine contains a large number of acid radicles which 
may be either organic or inorganic. Thus there are always present 
considerable amounts of chlorine, phosphorus pentoxide, sulphur 
trioxide, and (particularly in fasting experiments) /3-oxybutyric and 
other fatty acids. By using Folin's titration method, it has been possi- 
ble to determine the "total acidity." For a clear understanding of 
the quantitative relationships of these various acid radicles, direct 
determinations were made of the total chlorine, phosphorus pentoxide, 
total sulphur, and /3-oxybutyric acid. 

Chlorine. 

Owing to the general satisfaction with the standard Volhard method, 
chlorine has perhaps received more attention than any other of the 
inorganic constituents of the urine. Recognizing the importance of 
the determination of this constituent, especially in view of the emphasis 
laid upon the supposed excess storage of chlorine in the body under 



URINE. 



269 



normal conditions, we made duplicate determinations of the chlorine in 
the urine for each day of the fast. The chlorine excretion in fasting 
experiments has always had a particular interest, as the chlorine curve 
almost invariably follows a fairly regular course. It is furthermore 
important as presumably demonstrating whether or not the fast is 
a true one, for it is commonly supposed that unless the food sur- 
reptitiously taken is pure fat or pure carbohydrate, it is practically 
impossible for a subject to break his fast without almost immediately 
affecting the chlorine excretion. On the other hand, it was found, 



Table 31. — Chlorine (CI) excreted in urine 


daily by fasting 


r subjects. 




Day of fast. 


L. 


Succi. 


5 

O 


i 

« 


B 

m 
O 


< 

GO 


6 

6 

a 
I 

E 


J 
% 

a 
Z 


1 


> 


g 

a 

3 

w 


Last food day . . . 


gm. 


gm. 
2 6.322 

1.350 
.539 

1.155 
.848 
.817 
.840 
.800 
.736 
.550 
.513 
.332 
.405 
.230 
.119 
.137 
.113 
.130 
.258 
.298 
.311 
.234 
.216 
.219 
.235 
.204 
.118 
.139 
.239 
.428 
.688 


gm. 
4.51 
4.65 
2.86 
2.00 
.47 
.55 
.38 
.37 
.25 
.44 
.47 
.64 
.86 
.62 
.50 
.58 
.47 
.80 
.67 
.40 
.71 
.56 


gm. 

4.792 

3.908 

2.212 

1.799 

1.198 

1.092 

1.044 

.973 

.700 

.702 

.412 

.434 

.567 

.532 

.497 

.436 

.403 

.322 

.306 

.217 

.233 


gm. 


gm. 


gm. 
5.432 
1.606 
2.303 
1.7 
1.548 
1.396 
1.088 

.95 

.814 
1.104 

.62 


gm. 
6.7 
3.2 
2.0 
1.5 
1.3 

1.0 

.84 
.69 

.39 
.30 

.18 


gm. 

7.51 

2.99 

1.73 

3.66 

1.90 

.38 

.30 

.32 

1.15 

1.32 

1.07 

.98 

1.29 

.85 

.68 


gm. 
4.11 
1.45 
1.34 
.62 
.25 

!39 
.42 


1st 


3.77 
1.02 
.79 
.59 
.41 
.40 
.55 
.32 
.31 
.2.8 
.36 
.31 
.32 
.26 
.16 
.14 
.12 
.15 
.16 
.15 
.18 
.21 
.18 
.10 
.18 
.16 
.16 
.14 
.12 
.14 
.13 
.23 
.26 
8 .18 


9.03 

3.21 

1.55 

1.48 

1.18 

1.29 

1.11 

1.12 

1.21 

.87 

.92 

.82 

.58 

.65 

.51 

.45 

.58 

.44 

.67 

.42 

.42 




2d 


3d 


4th 


6th 


6th 


7th 


8th 


9th 


10th 


11th 


12th 






13th 






14th 






.24 


16th 






16th 












17th 












18th 












19th 












20th 












21st 












22d 












23d 








0.30 
.29 
.20 
.21 
.15 
.19 
.21 
.33 










24th 
















25th 
















26th 
















27th 
















28th 
















29th 
















30th 
















31st 
















1st food day .... 

2d food day 

3d food day 











































































Reported by the investigators as NaCl, but converted to chlorine for purposes of comparison . 
2 Average of 6 days before fast began. 
'Determined in the urine for about 22 hours. 



270 A STUDY OF PROLONGED FASTING. 

in the experiment with L., that when the subject took considerable 
amounts of food on the 3 days following the fast, the excretion of chlor- 
ine was but slightly affected, certainly not enough to be considered as 
proof that food had been taken. The character of the diet on these 
food days easily explains this absence of influence upon the excretion 
of chlorine. 

The determinations of the chlorine excretion were made under Dr. 
Peters's supervision by Mr. W. F. O'Hara according to the Volhard 
method, the excess of silver nitrate added being determined in a filtered 
portion of the urine. These values are compared in table 31 with the 
values obtained for several other subjects in long fasting experiments. 

With L. there was a large excretion of chlorine on the first day of the 
fast, doubtless from the food previously taken. This was followed by 
a marked fall, even on the second day, this decrease continuing almost 
regularly until the fifteenth day, when the excretion reached a new 
minimum level. It subsequently fluctuated slightly until the end of 
the fast. On the 3 food days there was a slight increase over the latter 
part of the fasting period. 

An intelligent comparison of the values found for L. with those found 
for other fasting subjects is somewhat difficult, owing to the facts that 
frequently the basis upon which the chlorine is reported is somewhat 
obscure and that some of the subjects, Succi in particular, were accus- 
tomed to drink water containing more or less chlorine. Apparently 
the chlorine excretion in complete inanition varies widely with different 
individuals, for it will be seen at once, by inspection of the values in 
table 31, that the excretion found for L., particularly in the first part 
of the fast, was lower than that found for any other subject except for 
a few days with Succi 1 in the Naples fast, for 3 days with Tosca, and for 
4 days with S. A. B. In the latter part of the long fasts, however, 
there is more of a tendency toward uniformity, although the values 
for Succi at Naples and at Vienna, and those for the latter portion 
of Tosca's fast, are much higher than those found with L. There is 
a general tendency shown with all of the subjects for the excretion to 
decrease gradually until the fifteenth day, but not so rapidly as was 
found for our subject. The observations of Brugsch on Succi at Ham- 
burg give values that agree well with those found for L. 

Relationship between chlorine excretion and preformed water lost. — A 
critical examination of the tables in the report of the earlier fasting 
experiments 2 shows a rather interesting relationship between the excre- 
tion of chlorine and the loss of preformed water from the body. For 
a long time we have been at a loss to explain the marked variations in 
the absolute amounts of chlorine excreted on different days of a fast 

*I have been unable to obtain the original chlorine data in the study made of Succi's urine by 
Koranyi, published in Orvosi hetilap, 1894, Nos. 39-40. See autoreferat, Maly, Jahrb. d. Tier- 
Chemie, 1894, 24, p. 268. Koranyi also studied the depression of the freezing-point. 

Benedict, Carnegie Inst. Wash. Pub. 77, 1907, table 216, page 415, and table 229, page 469. 



URINE. 



271 



by different subjects, there being almost no uniformity; some subjects, 
as Tosca, Cetti, and Succi in Vienna, excreting much larger amounts 
than others. In a relatively few fasting experiments, both the loss of 
preformed water from the body and the chlorine excretion have been 
determined, thus supplying data which permit comparison. Such a 
comparison is made in table 32, which gives the amount of water taken, 

Table 32. — Preformed water eliminated from the body and accompanying excretion of chlorine 
(CI) in experiments with fasting subjects. 



Subject. 


Day of 
fast. 


Water 
con- 
sumed. 

A 


Chlorine 
(CI) in 
urine. 

B 


Preformed water 
lost. 1 


Water in urine. 


Total. 1 
C 


Per 

gram of 
chlorine. 

(C-S-B) 

D 


Total. 
E 


Per 

gram of 

chlorine. 

(b-hb) 

F 


L 


1st 

2d 
3d 

4th 

1st 

2d 

1st 

2d 

1st 

2d 

1st 

2d 

1st 

2d 

1st 

2d 

1st 

2d 

1st 

2d 

1st 

2d 

1st 

2d 


gm. 

720 

750 

750 

750 

2082 

2747 

1973 

1729 

2048 

1593 

783 

340 

133 

206 

291 

194 

858 

1093 

1467 

884 

705 

708 

115 

357 


gm. 
3.77 
1.02 
.79 
.59 
1.63 
.47 
1.45 
1.34 
5.29 
1.67 
2.92 
3.62 
8.90 
4.03 
3.88 
2.79 
3.45 
6.71 
.52 
.63 
4.58 
1.46 
5.86 
1.81 


gm. 
585 
448 
350 
225 
377 
341 

-334 

273 

736 

713 

471 

623 

1500 

1132 

540 

777 

47 

991 

-142 
286 
744 
231 
822 
356 


gm. 
155 
439 
443 
381 
231 
726 

-230 
204 
139 
427 
161 
172 
169 
281 
139 
278 
14 
148 

-273 
454 
162 
158 
140 
197 


gm. 

630.2 

436.6 

530.6 

674.4 

2225.8 

2928.2 

1469.6 

1839.9 

2528.2 

2122.9 

996.1 

810.0 

1105.8 

734.8 

621.3 

783.2 

538.6 

1740.1 

1159.2 

1012.4 

1145.1 

642.4 

599.6 

492.2 


gm. 

167 

428 

672 

1143 

1366 

6230 

1014 

1373 

478 

1271 

341 

224 

124 

182 

160 

281 

156 

259 

2229 

1607 

250 

440 

102 

272 


S.A.B., Exp. 73... 

Exp. 75 . . . 

Exp. 77. .. 
H.E. S., Exp. 79. .. 

C. R. Y., Exp. 80. . . 
A. H. M., Exp. 81 . . 
H. C. K., Exp. 82 . . 
H. R. D., Exp. 83 . . 
N.M.P.,Exp.85.. 

D. W., Exp. 89. . . . 



1 Preformed water other than that resulting from the disintegration of flesh and fat. (Column 
j, table 62, and Carnegie Institution of Washington Pub. 77, table 229, column c). Since the 
chlorine in flesh is not a large proportion of the total chlorine, the water of flesh and fat is purposely 
omitted in this discussion. 

the chlorine excreted in the urine, the preformed water lost from or stored 
in the body, 1 the preformed water lost per gram of chlorine excreted, 
the water in the urine, and the water in the urine per gram of chlorine 
excreted during fasting. These factors are given not only for the 
first 4 days of the experiment with L., but also for the first 2 days for 

1 For discussion of this factor, see page 408. 



272 A STUDY OF PROLONGED FASTING. 

a number of short fasting experiments made with eight subjects at 
Wesleyan University. 

It is conceivable that variations in the chlorine excretion might be 
due, in part at least, to a flushing out of the body, and hence we should 
expect to find the chlorine excretion varying to a certain degree with the 
variations in the volume of urine or of the water taken. An examina- 
tion of the figures in the last column of table 32 (column f) shows, 
however, that the water in the urine per gram of chlorine excreted 
undergoes wide variations, ranging from 102 grams to 6,230 grams, with 
no obvious average value. 

If we consider the body as losing regularly not only carbohydrate, fat, 
and protein from its original store of substance at the beginning of 
the fast, but also losing regular amounts of preformed water, i. e., water 
of flesh and fatty tissue (see note to table 32), and water existing in 
the fluids of the body, we can see that the importance of the determina- 
tion of the preformed water lost is much greater than would at first 
appear. The values for the preformed water lost by these fasting sub- 
jects are given in column c of table 32, and for the preformed water 
lost per gram of chlorine in column d. An examination of the values 
for the preformed water lost per gram of chlorine shows that there is 
at least a semblance to regularity. This is more clearly seen if the first 
day of the experiment is omitted, as may properly be done, since it is 
natural to suppose that the chlorine excretion on the first day may 
have been influenced by the previous diet. Excluding the values for 
the first day, then, we find that the remaining values range from 726 on 
the second day of experiment 73 with S. A. B. to as low as 148 on the 
second day of experiment 82 with H. C. K. Aside from the very high 
value found in experiment 73, it will be seen that there is a fairly close 
agreement between the chlorine excreted and the preformed water lost 
from the body. It would thus appear that, in the discharge of this 
water, there is excreted simultaneously an amount of chlorine approx- 
imately proportional to the total preformed water lost. It is obvious 
that all of the factors involved in the determination of the preformed 
water lost from the body are such as to make the absolute values of 
some of the determinations problematical, and yet we believe that as 
a whole there is sufficient agreement here to indicate some approximate 
relationship between the chlorine and the preformed water lost from 
the body. 

Source of chlorine excreted. — The exact source of the chlorine excreted 
in the first days of fasting is by no means certain, but it is clear that 
the small amounts excreted, in the urine after the first few days of 
fasting, correspond to the usual percentage of chlorine commonly con- 
sidered as belonging to human flesh. Thus Katz, 1 whose analyses 
have been considered as remarkably accurate, maintains that human 

x Katz, Archiv f. d. ges. Physiol., 1896, 63, p. 1. 



URINE. 273 

muscle contains 0.07 per cent chlorine. Furthermore, Magnus-Levy's 1 
analysis of the flesh of a suicide agrees remarkably well with the values of 
Katz. If we use this factor 0.07 for computing the chlorine in the flesh 
katabolized by the subject L. as recorded in column q, table 61 (page 
403), the values found would be approximately the amounts of chlorine 
actually excreted. For instance, on the twenty-fifth day, there were 
235 grams of flesh katabolized, the excretion of chlorine being somewhat 
larger than the average for this part of the fast. Applying the factor 
of Katz, we find that 0.07 per cent of 235 grams would give 0.165 gram 
of chlorine, while the amount actually excreted on that day as shown 
by analysis was 0.18 gram. It is clear, therefore, that, at least in the 
later stages of inanition, chlorine is derived for the most part from dis- 
integrated muscle substance. The large storage of salt in the skin, 
which was noted by Wahlgren 2 and subsequently further studied by 
Padtberg, 3 and Scholz and Hinkel, 4 must therefore have been rapidly 
depleted during the first days of the fast. In any event, the total 
amount of chlorine involved throughout the whole 31 days of our 
fasting experiment was but 12.27 grams, an amount so small as to cast 
a doubt upon the theory that there is an excess 5 of chlorine stored in 
the body. 

Phosphorus. 

Since phosphorus has so intimate a relationship with both the mineral 
and the organic constituents of the body, observations have been made 
of the amounts present in the fasting urine of a large number of subjects. 
The phosphorus in the urine of L. was determined by Mr. W. F. 
O'Hara under the supervision of Dr. Peters, for each day of the fast 
and for the following food days, by titration with uranium acetate. 
Usage is followed here in expressing the values as phosphorus pentoxide 
instead of as phosphorus, although the inconsistency of expressing 
the elements in terms of their compounds is obvious. The absolute 
amounts of phosphorus pentoxide determined in these urines, together 
with those found in other long fasting experiments, are given in table 33. 

The values obtained for L. show an increasing amount for the first 
4 days and thereafter a very regular decrease for the remainder of the 
fast. The maximum amount, 2.90 grams of phosphorus pentoxide, 
was observed on the fourth day and the minimum amount, 1.32 grams, 
on the last day. While in the first part of the fast the values for L. 
are exceeded by those for Cetti and Beaute* and approximately equaled 
by those for S. A. B., in the latter part of the fast they are measurably 

Magnus-Levy, Biochem. Zeitschr., 1910, 24, p. 363. 

*Wahlgren, Archiv f. exp. Path. u. Pharm., 1909, 61, p. 97. 

'Padtberg, Archiv f. exp. Path. u. Pharm., 1910, 63, p. 60. 

<Scholz and Hinkel, Deutsch. Archiv f. klin. Med., 1913, 112, p. 334. 

*For a discussion of this point, see Magnus-Levy, Physiologie des Stoffwechsels, von Noorden's 
Handbuch der Path, des Stoffwechsels, Berlin, 1906, 1, p. 451; Munk, Archiv f. Path. Anat. u. 
Physiol., 1893, Supp. 131, p. 146; and Morawitz, Oppenheimer's Handbuch der Biochemie, Jena, 
1910, 4 (2), p. 282. 



274 



A STUDY OF PROLONGED FASTING. 



higher than those recorded for any other subject. The only other fast 
in which the phosphorus was determined, and which extended over 
so long a period as that for our subject, was Succi's fast in Florence, 
but the amount excreted by this subject was considerably less than 
that found for L. The values obtained by Brugsch on Succi in the 

Table 33. — Phosphorus (P2O6) eliminated in urine daily by fasting subjects. 



Day of fast. 


L. 


Succi. 


"■+3 

■♦a 

6 


<3 

1 

S 


c3 

8 

O 


PQ 
< 
00 


i 

a 

% 


! 


6 

1 


03 


t* 

1 




gm. 


gm. 


gm. 
1.90 
1.78 
1.82 
1.95 
1.46 
2 64 
2.47 
2.32 
1.48 
1.49 
1.23 
1.22 
1.98 
1.11 
1.14 
1.33 
1.50 
1.02 
1.36 
1.02 
1.19 
1.11 


gm. 

1.792 

2.499 

1.559 

1.528 

1.662 

2.100 

1.561 

1.678 

1.158 

.841 

.662 

.518 

.769 

.879 

.428 

.465 

1.162 

1.079 

.725 

.610 


gm. 


gm. 


gm. 
2.76 
2.597 
2.925 
3.289 
2.974 
2,871 
2.667 
2.663 
1.722 
2.065 

.948 


gm. 
4.14 
2.26 
2.93 
2.98 
2.91 

2.37 
1.84 
1.89 

1.60 
1.54 
1.55 


gm. 
2.670 
1.550 
1.830 
2 654 
2.934 
1.749 
1.069 

.713 
1.658 
1.702 
1.461 
1.097 
1.312 
1.114 

.869 


gm. 
2.318 
1.431 
2.256 
2.055 
2.406 
2.078 
2.071 
2.081 


1st 


1.66 
2.48 
2.51 
2.90 
2.64 
2.33 
1.84 
1.84 
2.13 
1.97 
1.95 
1.70 
1.95 
1.86 
1.47 
2.04 
1.99 
1.86 
1.75 
1.47 
1.60 
1.57 
1.62 
1.55 
1.53 
1.49 
1.41 
1.35 
1.46 
1.39 
1.32 
.74 
.31 
*.21 


1.9W 

2.051 

2.090 

2.120 

2.394 

2.150 

1.865 

1.601 

1.360 

1.246 

1.420 

1.012 

.363 

.996 

1.029 

1.077 

1.218 

1.005 

.953 

.875 

.747 

.718 

1.049 

.790 

.592 

.783 

.861 

.945 

.789 

1.019 


2.98 
2.75 
2. 52 
2.54 
2.51 
2.27 
2.13 
2.31 
2.40 
1.68 
1.41 
1.35 
1.04 

.99 
1.32 

.876 
1.34 

.86 
1.14 

.67 

.64 




2d 

3d 

4th 

6th 

6th 


7th 


8th 


9th 


10th 


11th 


12th. . 






13th. . . 






14th 






1.25 


15th. . 






16th. 












17th. . . 












18th. . . 












19th. . 












20th. . 












21st 












22d . . 












23d 








0.96 

1.062 
.980 
.900 

1.056 
.901 
.754 

1.545 










24th 
















25th 
















26th 
















27th. . 
















28th. . 
















29th. . 
















30th. 
















31st 
















1st food day. . 
2d food day . . 
3d food day. . 











































































♦Determined in urine for about 22 hours. 



Hamburg fast are also measurably less than those reported for our 
fasting subject. 

Relationship between phosphorus and total nitrogen. — Owing to the 
intimate relationship between phosphorus and the organic tissues of the 
body, particularly muscle, the ratio between phosphorus and total 



URINE. 



275 



nitrogen has frequently been computed for fasting experiments. The 
well-known determinations of phosphorus pentoxide and nitrogen in 
muscle show that for each gram of phosphorus pentoxide there should 
be 6.6 grams of nitrogen. The ratios between nitrogen and phos- 
phorus pentoxide have been computed, not only for the fasting experi- 
ment with L., but likewise, in so far as the data permit, for those of 



Table 34.- 


—Ratio 


of nitrogen to phosphorus (^r^r-) 


in urine of fasting subjects. 


Day of fast. 


L. 


Succi. 


Cetti. 


Beaute. 


Tosca. 


S. A. B. 


Flor- 
ence. 


Rome. 


Vienna. 


Ham- 
burg. 








5.10 
3.57 
5.88 
5.68 
5.09 
4.77 
6.03 
5.11 
7.03 
7.55 
8.63 
9.54 
6.65 
5.44 
10.30 

6.77 
2.86 
3.76 
5.27 
5.66 






4.89 
6.22 
4.30 
3.99 
4.17 
3.73 
3.79 
4.09 
5.17 
5.24 
10.00 


3.97 
4.65 
4.90 
4.60 
4.71 

4.54 
5.25 
5.03 

5.23 
5.51 
5.65 


5.24 
5.65 
4.58 
4.04 
3.20 
4.50 
7.23 
8.57 
4.64 
4.32 
4.65 
5.60 
5.31 
5.04 
4.70 


8.41 
8.55 
5.52 
6.34 
4.83 
5.23 
5.19 
4.87 


1st 


4.28 
3.39 
4.52 
4.09 
3.94 
4.37 
6.32 
6.58 
5.04 
5.10 
5.26 
5.96 
5.31 
5.61 
5.76 
4.70 
4.43 
4.45 
4.78 
5.23 
4.96 
4.94 
4.51 
5.26 
5.10 
6.29 
5.72 
5.64 
5.16 
5.63 
5.26 
6.53 
12.29 
13.10 


*7.87 
5.91 
7.30 
6.64 
5.90 
5.18 
5.53 
5.85 
6.29 
5.96 
6.11 
7.79 

10.63 
5.89 
6.50 
5.62 
5.57 
6.97 
5.81 
5.51 
6.72 
4.90 
4.99 
7.73 

11.23 
7.11 
6.85 
6.52 
5.69 
7.14 


5.70 
4.07 
4.19 
4.25 
4.46 
4.85 
4.13 
4.22 
4.19 
4.24 
4.48 
5.07 
4.94 
4.71 
3.83 
4.83 
4.03 
4.19 
5.00 
4.93 
4.41 




2d 


3d 


4th 


5th 


6th 


7th 


8th 


9th 


10th 


11th . 


12th. . 






13th . 






14th 






6.22 


15th . 






16th. . 












17th 












18th 












19th 












20th 












21st. . 












22d 












23d 






6.08 
6.04 
6.40 
6.87 
5.97 
4.93 
5.56 
5.45 










24th . 














25 th 














26th 














27th . 














28th. . 














29th. . 














30th 














31st . 














1st food day. . . 
2d food day. . . 
3d food day .... 



































































♦The ratios shown in this column have been obtained by means of the nitrogen as corrected 

by Munk. (See table 26, p. 249.) 

i 

the earlier fasting experiments reported in table 33. These ratios are 
given in table 34. 

Considering the fasting values for L., we find that in no case do they 
reach the theoretical relationship found with muscle, namely, 6.6. The 
highest value was 5.96 on the twelfth day and the lowest value was 



276 A STUDY OF PKOLONGED FASTING. 

3.39 on the second day of the fast. Furthermore, the figures show no 
definite increment in the ratio as the fast progressed. While it is true 
that values less than 5 are not found on the last 8 days of the fast, 
nevertheless there is a period — that between the seventh and the fif- 
teenth days — when the values again all lie above 5, while between the 
fifteenth and the twenty-fourth days values as low as 4.4 are found. 

In examining the values for the other fasts, we find several which 
show distinctly abnormal values for the first day, i. e., 7.87 for the 
first day of Succi's Florence fast, and 8.55 for the first day of the fast 
of S. A. B., but this would naturally be expected. On the other hand, 
after the first day the values should lie well below 6.6 and any above 
this are open to suspicion. The ratios established by the Freunds for 
the Vienna fast of Succi are remarkably constant, as are those of 
Brugsch for a later fast of Succi in Hamburg. Perhaps the greatest 
uniformity in ratios is shown by Cathcart's subject Beauts, these 
ratios gradually and steadily increasing as the fast progressed. The 
values for Succi in the Florence fast are somewhat vitiated by the 
uncertainty in the nitrogen determinations, although the values were 
computed on the basis of nitrogen as corrected by Munk. 

Source of phosphorus excreted. — Comparing these ratios, particularly 
those for L., with the theoretical relationship with flesh, we find in 
all of the experiments a tendency toward a much larger excretion of 
phosphorus pentoxide in its relation to nitrogen than occurs in the 
ordinary composition of flesh. The possible sources of phosphorus in 
the body other than the flesh are of course the nucleins and, above all, 
the mineral matter of the bony structure. It is now the consensus 
of opinion that the disturbance in the relationship between the phos- 
phorus pentoxide and nitrogen in fasting experiments is due exclusively 
to the material draft upon the bony structure as the fast progresses. 1 
This was clearly set forth by Munk in his discussion of the experi- 
ments with Cetti and Breithaupt, but as these were short experiments 
Munk frankly stated that he expected to find that the ratio would 
become smaller and smaller as the fast progressed. The ratios in the 
31-day fast with L. do not, however, become smaller as the fast con- 
tinued, but on the other hand tend to become higher in the last week 
than at any other time. Brugsch has already commented upon the 
very high values found in the last week of his study with Succi. 

During the fast with L., 277.32 grams of nitrogen 2 were excreted, 
corresponding to 8,319.6 grams of flesh katabolized. If we assume 
that this flesh had normally combined with it 0.5 per cent of phos- 
phorus pentoxide, the total amount combined with the katabolized 
flesh would be equal to 41.6 grams. Since 56.63 grams of phosphorus 

x Wellman (Archiv f. d. ges. Physiol., 1907-1908, 121, p. 508) believes that the calcium and 
phosphorus losses found by him conform with Munk's conclusion that there is considerable 
loss of calcium and phosphorus from the bones during starvation. 

2 This amount does not include the nitrogen excreted through the skin (see table 22). 



URINE. 277 

pentoxide were excreted during the fast, it will be seen that there was 
distinctly an excess excretion, amounting to 15 grams of phosphorus 
pentoxide for the whole fast. This was undoubtedly derived — in large 
part, at least — from the bones. It was hoped that the present-day 
technique of Roentgen photography would show any material attacks 
upon the bony structure and possible depletion of calcium, but the 
excellent series of X-ray photographs taken by Dr. Francis H. Williams, 
of the Boston City Hospital, did not indicate this. On the other hand, 
when one considers the large storage of calcium in the body and the 
relatively small draft, an ocular indication of such a draft which could 
be measured could hardly be expected. 

The exact apportionment of the phosphorus pentoxide between 
muscle and bone is not permissible with the experimental data at 
present in our hands. In all probability the amount of phosphorus 
pentoxide actually drawn from the skeleton was considerably more 
than 15 grams. 

Since after the first few days there was no material increase in the 
uric-acid excretion as the fast progressed, there was doubtless no direct 
attack upon the nucleins, and thus the increase in the phosphorus 
pentoxide could not have been derived from that source. 

Sulphur. 

Sulphur as an integral component of protoplasm is oxidized and ex- 
creted in the urine in several forms: first, as sulphates; second, in the 
form of conjugated sulphates, or sulphuric acid combined with organic 
radicles; and finally, as so-called neutral or unoxidized sulphur. The 
apportionment of the total sulphur output among these various com- 
ponents has been studied during fasting in considerable detail, both 
in the laboratory of Wesleyan University and by Cathcart. In the 
series of experiments carried out on L., it was impossible to separate 
the sulphur, owing to the lack of experimental material, and only the 
total sulphur was determined. These determinations were personally 
made by Dr. A. W. Peters, the constancy in the results testifying to 
his technical skill. 

It has been the custom of many writers to report the sulphur excre- 
tion as sulphuric acid or sulphur trioxide, but the values given for L. 
in table 35 represent the daily excretion of total sulphur. For com- 
parison, the sulphur excretion has been computed from the results 
given by other observers for fasting subjects, and these values are 
included in the table. Two difficulties arise in comparing our results 
with those obtained with other subjects. In the first place, frequently 
only the total sulphuric acid was determined and the organic sul- 
phur was not included. Secondly (in the case of Succi, at least), the 
subject often drank mineral water containing measurable amounts of 
sulphates. 



278 



A STUDY OF PROLONGED FASTING. 



With our subject L., the sulphur excretion followed a course not unlike 
that of the nitrogen excretion, there being a slightly increasing amount 
excreted for the first 3 days and thereafter an almost continuous 
decrease until the end of the fast. The maximum amount (0.68 gram) 

Table 35. — Total sulphur (S) excreted in urine daily by fasting subjects. 



Day of fast. 


L. 


Succi. 




tt 














Florence. Na 


pies. 


Rome. Vie 


nn 


a. 


L. 1H auu'. 




Last food day. . 
1st 


gm. 

0.46 
.61 
.68 
.67 
.65 
.65 
.62 
.64 
.66 
.61 
.62 
.62 
.62 
.60 
.60 
.69 
.53 
.54 
.55 
.61 
.51 
.50 
.61 
.49 
.49 
.64 
.52 
.53 
.52 
.52 
.49 
.39 
.22 

».36 


gm. 1 g 
1 
0.75 

.72 

.75 

.71 

.68 

.59 

.65 

.49 

.43 

.42 

.45 

.43 

.30 

.33 

.28 

.26 

.32 

.21 

.22 

.24 

.21 

.19 

.29 

.35 

.28 

.19 

.17 

.19 

.14 

.24 


.04 
.76 
.48 
.72 
.60 
.58 
.40 
.49 
.48 
.44 
.68 
.30 
.79 
.38 
.29 
.20 
.31 
.70 
.66 
.35 
.46 
.61 


gm. g 
0.632 
.690 1 
.277 
.346 
.249 
.312 
.187 
.276 
.258 

.272 ! 

.120 
.130 

.251 
.207 
.117 


m. 

21 

5: 

6- 

r>( 

4C 

k 

3f 

2i 

3( 

2; 


gm. 

t 0.99( 

1 

! .921 

t 

.71] 

\ .78] 

.831 
.61( 

> 


( 
1. 
) 

. 

! 

L 
I 


im. 

33 

614 

934 

801 

856 

712 
644 
615 

556 
570 
577 

536 


f 

1 


im. 
36 
62 
67 
75 
72 
67 
66 
62 


2d 


3d 


4th 


6th 


6th 


7th 


8th 


9th 






10th 


11th 


12th 


i 

) '. 




13th 


14th 


15th 


16th 






17th 


18th 


19th 


20th 


21st 


22d 


23d 


24th 


25th 


26th 


27th 


28th 


29th 


30th 


31st 


1st food day. . . 
2d food day. . . . 
3d food day 



x Probably inorganic and ethereal. Given by the investigator as H2SO4, but converted to S 
for purposes of comparison. 

2 Probably inorganic and ethereal. 
3 Determined in urine for about 22 hours. 

was excreted on the third day of the fast, and the minimum amount 
(0.49 gram) was found on the twenty-fourth, twenty-fifth, and thirty- 
first days. 

The most carefully determined results for the other subjects are 
undoubtedly those for Cathcart's subject Beaute\ Here again we find 
a decrease in the sulphur excretion as the fast progresses, the values for 



URINE. 



279 



the entire fast being not unlike those found for L. This is likewise true 
for the results of the observations on S. A. B. 

Relationship between total nitrogen and total sulphur. — The total 
sulphur excretion has a special significance in that it is so intimately 
combined with the protoplasm that it is frequently suggested as an 
index of the total muscle katabolized. Since there is a relatively 
constant relation between the nitrogen and sulphur in muscle, i. e., 

Table 36. — Ratio of nitrogen to total sulphur (— ) in urine of fasting subjects. 



Day of fast. 


L. 


Succi at , 
Vienna. 


3etti. B 


eau 


te. S. A. B. 


Last food day 

1st 


15.4 
13.8 
16.7 
17.7 
16.0 
15.7 
15.8 
16.0 
16.3 
16.5 
16.6 
16.3 
16.7 
17.4 
16.9 
16.2 
16.6 
15.3 
15.2 
15.1 
15.5 
15.5 
14.3 
16.6 
15.9 
14.6 
15.5 
14.4 
14.5 
15.1 
14.2 
12.4 
17.3 
7.6 


13 .3 
19.0 
20.3 
16.9 



19.7 
15.7 
21.2 

] 

] 

19.7 
18.0 

14.6 
15.0 

14.3 


13.7 

L4.*2 

L5.1 

L3.9 

... 

13. 

L5.3 

; 

... 

] 


L2. 
L7. 

L5. 
L7. 

LG. 

L5. 
15. ( 

L5.< 

15. ( 
14. < 
15. J 

L4.\ 


2 14.3 
I 19.6 
J 18.6 
L 17.4 
) 16.1 
16.3 
I 16.3 
) 16.3 
I 


2d 


3d 


4th 


5th 


6th 


7th 


8th 


9th 


) 
I 




10th 


11th 


12th 


13th 


14th 


15th 


16th 


17th 


18th 


19th 


20th 


21st 


22d 


23d 


24th 


25th 


26th 


27th 


28th 


29th 


30th 


31st 


3d food day 



about 13.3 grams to 1 gram of sulphur, the relationship between the 
excretions of total nitrogen and total sulphur is worthy of note. The 
nitrogen-sulphur ratio has been computed not only for the subject L., 
but for a number of the other fasting subjects; these computed ratios 
are given in table 36. But one ratio for Succi is included in this table, 
that for the Vienna fast, in which the observations were made by the 
Freunds, for while it was permissible to include the observations of the 



280 A STUDY OF PROLONGED FASTING. 

sulphur excretion in the other fasts in table 35, it was not permissible 
to compute the ratio between the nitrogen and the sulphur, since there 
were undoubtedly errors in the nitrogen determinations and probably 
likewise in those for the sulphur excretion. 

The ratios found for L. show at first a slight increase, rising on the 
fourth day to a maximum of 17.7. They then decrease with consider- 
able regularity, the lowest value, 14.2, being on the last day of the fast. 
In general, the ratios remain within very narrow limits. These values 
are again not unlike those found by Cathcart with Beauts, which 
showed a tendency to decrease as the fast progressed. The same ten- 
dency is shown by the values for S. A. B., although at no time during 
the 7-day fasting experiment did the ratio fall below 16.1. 

All of the values found for L. show a somewhat higher excretion of 
nitrogen than would normally accompany the amount of sulphur ex- 
creted; it would appear, therefore, that there was a disintegration of 
sulphur-poor and nitrogen-rich substance other than muscle. Through- 
out all of the observations, it has been noted that on certain days there 
was always a marked lowering in the excretion of the urinary compo- 
nents. This lowering may be due either to an actual decrease in the 
katabolism on that day or to a possible loss of urine. The handling, 
sampling, and preservation of the urine were so strictly controlled that 
it would seem impossible for such a loss to occur. It is conceivable, 
of course, that the subject may not have emptied the bladder com- 
pletely in the morning, but a compensating increase in the urine of 
the next day would then be expected, which was not observed. If we 
examine the values for sulphur, total creatinine, nitrogen, chlorine, 
and phosphorus, we find that there was a distinct lowering in the 
amount excreted on the fifteenth day of the fast, with occasionally a 
slight indication of a compensating increase on the sixteenth day. 
This points toward a loss of urine which we are as yet unable to 
account for. As a matter of fact, no disturbance in the nitrogen- 
sulphur ratio was found on the fifteenth day and, indeed, such dis- 
turbance was not expected. 

On the other hand, we note a very definite increase in the sulphur- 
nitrogen ratio from 14.3 on the twenty-third day to 16.6 on the twenty- 
fourth day. An examination of the values for the total nitrogen 
excretion shows that here also there was a marked increase, thus indi- 
cating the disintegration of a nitrogen-rich and a sulphur-poor sub- 
stance other than muscle. The general course of the values for sulphur 
and total creatinine show a striking similarity, indicating that the 
katabolism resulting in the excretion of total creatinine is accompanied 
by an excretion of sulphur. If we are to accept Folin's view that the 
total creatinine is an admirable index of the total katabolism of tissue, 
we may then conversely state that the sulphur is likewise an index. 



URINE. 



281 



Total Acidity. 

When the conditions are such that acidosis may be expected to 
develop, as in fasting, a determination of the total acidity of the urine 
is of special value. Accordingly the total acidity of the urine in this 
experiment was determined for each day of the fasting period and also 
for the 3 days following when food was taken. Under the direction 
of Dr. A. W. Peters, the determinations were made by W. F. O'Hara, 
according to the method of Folin, 1 in which potassium oxalate was used 
and the titration was carried out with 25 c.c. of urine. The values are 
given in table 37, expressed in terms of cubic centimeters of deci-normal 
sodium hydroxide solution. 

Table 37. — Total acidity (— NaOH) of urine of fasting subjects. 



Day of fast. 


L. 


Beaut6. 


Day of fast. 


L. 


Succi at 
Hamburg. 


Last food day . . 
1st 


c.c. 
*409 
285 
499 
658 
655 
570 
467 
337 
399 
415 
376 
365 
297 
362 
1 332 
313 
505 
485 


c.c. 
582 
378 
640 
687 
604 

454 * 

358 

344 

280 
256 
212 

228 


18th 


c.c. 
383 
347 
264 
296 
301 
287 
328 
303 
289 
284 
253 
268 
263 
227 
139 
56 
3 59 


c.c. 

2 384 
295 
263 
300 
283 
265 
156 
103 


19th 


2d 


20th 


3d 


21st 


4th 


22d 


5th : 

6th 


23d 


24th 


7th 


25th 


8th 


26th 


9th 


27th 


10th 


28th 


11th 


29th 


12th 


30th 


13th 


31st 


14th 


1st food day. . 
2d food day. . 
3d food day . . 


15th 


16th 


17th 





1 Acidity of urine on third food day before the fast. 

2 The figures in this column were reported by the investigator as normal sulphuric acid, but 
are here proportionately increased for purposes of comparison. 
"Determined on urine for 22 hours. 

At the beginning of the fast, the acidity increased rapidly until the 
maximum of 655 c.c. was reached on the fourth day. It then showed a 
general tendency to decrease as the fast progressed, although occa- 
sionally unusually high values were found, as on the sixteenth and 
seventeenth days of the fast. In the 3 food days following the fast, the 
acidity dropped almost immediately to about 60 c.c. 

The fasting values best suited for comparison with the values for 
acidity found for L. are those determined by Cathcart on Beaute* and 
by Brugsch on Succi in Hamburg. Cathcart's figures for the total 
acidity in the 14-day fast of Beaute* show values ranging from a maxi- 

^olin, Am. Journ. Physiol., 1903, 9, p. 265. 



282 A STUDY OF PKOLONGED FASTING. 

mum of 687 on the third day of the fast to a minimum of 212 on the 
twelfth day of the fast. Following the fasting period the subject was 
given a starch-cream diet, and the acidity immediately dropped to 
approximately 100 c.c. 

From the twenty-third to the thirtieth days of the fast in Hamburg, 
Brugsch found with Succi values ranging from 384 c.c. on the twenty- 
third day to 103 c.c. on the thirtieth day — values which are not incom- 
parable with those found by us with L. 

It was impossible, owing to the insufficient supply of urine, to deter- 
mine the mineral acidity in the urine of our subject. 

jS-OXTBTJTTBIC AdD. 

In the earlier fasting experiments at Wesleyan University, there was 
strong evidence that a nitrogen-poor, carbon-rich substance was present 
in the urine in large amounts. This was shown by determinations of 
the carbon in the urine and the relationship between the carbon and 
nitrogen and calories. At that time the opinion was expressed that in 
all probability the material was jS-oxybutyric acid, but we were then 
unable to make the determinations in addition to the other analyses. 
Special effort was therefore made in this fasting experiment to deter- 
mine the /3-oxybutyric acid in the urine as accurately as possible, 
although the interesting paper of Brugsch, reporting the Hamburg 
fast with Succi, and more recently the paper by Grafe, 1 have left no 
doubt as to the nature of this excess non-nitrogenous material in the 
urine of a fasting man. 

Results of determinations. — Of the methods for the determination of 
/3-oxybutyric acid which were available at the time of this experiment, 
that of Black 2 was best fitted for our purpose. By this method, plaster 
of paris is first mixed with the acidulated, dried urine and the mixture 
is then extracted with ether, the /3-oxybutyric acid removed being 
determined with the polariscope. 3 The determinations were carried 
out by Miss Alice Johnson under the supervision of Dr. A. W. Peters. 

Since these determinations were made, a large amount of research 
on /3-oxybutyric acid has been carried out; in the light of the technique 
existing at the time of the experiment, however, the determinations, 
while admittedly having a relatively large error, are nevertheless suffi- 
ciently accurate to indicate that there was a material excretion of 
jS-oxybutyric acid throughout the fast. The results are recorded in 
table 38 (column f). 

On the second day of the fast only 0.5 gram of /3-oxybutyric acid was 
found; this was wholly in fine with what would be expected. Subse- 
quently no values less than 1.4 grams were found until the fast had been 

x Grafe, Zeitschr. f. phyaiol. Chemie, 1910, 65, p. 21. 
2 Black, Journ. Biol. Chem., 1908, 5, p. 207. 

'For a more explicit statement of this procedure, see Benedict and Joslin, Carnegie Inst. Wash. 
Pub. 136, 1910, p. 25. 



URINE. 



283 



concluded. On the first day with food, the /3-oxybutyric acid dropped 
at once to 0.8 gram. During the last week of the fast, the results 
obtained by the optical method showed a considerable amount of 
/3-oxybutyric acid present in the urine. 



Table 38. — p-oxybutyric acid excretea 


in urine in experiment with L. 




Date. 


Day of 

fast. 


Total 
nitrogen. 


Carbon. 


/3-oxybutyric acid. 


Computed 
normal 


Deter- 


Differ- 


Calcu- 
lated. 


Deter- 








excretion. 


mined. 


(c-b) 


/dX100\ 
^ 46.11 ' 


mined. 








(aXO.79). 










A 


B 


C 


D 


£ 


F 


1912. 




gm. 


gm. 


gm. 


gm. 


gm. 


gm. 


Apr. 14-15 


1st 


7.10 


5.61 


6.82 


0.21 


0.46 




15-16 


2d 


8.40 


6.64 


7.99 


1.35 


2.93 


6.5 


16-17 


3d 


11.34 


8.96 


10.35 


1.39 


3.01 


2.1 


17-18 


4th 


11.87 


9.38 


11.88 


2.50 


5.42 


3.5 


18-19 


5th.... 


10.41 


8.22 


10.69 


2.47 


6.36 


2.1 


19-20 


6th ... . 


10.18 


8.04 


10.42 


2.38 


6.16 


3.6 


20-21 


7th 


9.79 


7.73 


9.06 


1.33 


2.88 


2.8 


21-22 


8th.... 


10.27 


8.11 


10.30 


2.19 


4.75 


1.6 


22-23 


9th 


10.74 


8.48 


10.92 


2.44 


6.29 


3.5 


23-24 


10th 


10.05 


7.94 


9.92 


1.98 


4.29 


3.5 


24-25 


11th 


10.25 


8.10 


9.59 


1.49 


3.23 


1.4 


25-26 


12th 


10.13 


8.00 


9.05 


1.05 


2.28 


2.4 


26-27 


13th 


10.35 


8.18 


10.15 


1.97 


4.27 


4.2 


27-28 


14th 


10.43 


8.24 


9.95 


1.71 


3.71 


4.7 


28-29 


15th 


8.46 


6.68 


8.71 


2.03 


4.40 


1.6 


29-30 


16th 


9.58 


7.57 


11.39 


3.82 


8.28 


5.2 


Apr. 30-May 1 . . 


17th.... 


8.81 


6.96 


10.91 


3.95 


8.57 


3.6 


May 1-2 


18th 


8.27 


6.53 


9.65 


3.12 


6.77 


4.4 


2- 3 


19th.... 


8.37 


6.61 


9.56 


2.95 


6.40 


7.0 


3-4 


20th.... 


7.69 


6.08 


8.07 


1.99 


4.32 


4.4 


4-5 


21st 


7.93 


6.26 


8.69 


2.33 


5.05 


5.0 


5- 6 


22d 


7.75 


6.12 


8.40 


2.28 


4.95 


3.1 


6- 7 


23d 


7.31 


5.77 


7.25 


1.48 


3.21 


6.0 


7-8 


24th 


8.15 


6.44 


8.68 


2.24 


4.86 


6.9 


8- 9 


25th.... 


7.81 


6.17 


8.58 


2.41 


5.23 


4.4 


9-10 


26th.... 


7.88 


6.23 


8.56 


2.33 


5.05 


6.1 


10-11 


27th .... 


8.07 


6.38 


8.23 


1.85 


4.01 


4.0 


11-12 


28th 


7.62 


6.02 


7.73 


1.71 


3.71 


4.9 


12-13 


29th 


7.54 


5.96 


7.94 


1.98 


4.29 


5.6 


13-14 


30th .... 


7.83 


6.19 


7.95 


1.76 


3.82 


5.4 


14-15 


31st 


6.94 


5.48 


7.37 


1.89 


4.10 


4.5 


15-16 




4.83 
3.81 
»2.75 


3.82 
3.01 
2.17 


7.13 

4.28 


3.31 
1.27 


7.18 
2.75 


.8 

.5 

1.5 


16-17 




17-18 









determined in urine for about 22 hours. 

The determination of /3-oxybutyric acid by this method was so 
defective that it seems unwise to compute the amount of ammonia 
which would theoretically combine with the acid and to discuss any 
relationship arising therefrom. 

Indirect computation of amounts of p-oxybutyric acid excreted. — In 
addition to the determinations made by the Black method, the amounts 



284 A STUDY OF PROLONGED FASTING. 

of /8-oxybutyric acid excreted were also computed by an indirect 
method, using the relationship between the nitrogen and the carbon. 
The carbon and the nitrogen in the urine were determined for every- 
day of the fast and for the 3 food days following the fasting period. 
Normally there exists a relatively definite relationship between the 
carbon and the nitrogen of urine, a relationship which was determined 
by Munk on Breithaupt and Cetti as being not far from 1 part of 
nitrogen to 0.82 of carbon. This relationship was determined for L. 
for 2 days before the fast, and a ratio found, which may be termed 
"normal," of 1 to 0.79. We therefore computed the amount of carbon 
that would normally be excreted from the amount of nitrogen actually 
excreted, and deducted the result found from the total amount of 
carbon found in the urine. The excess carbon would be due, in all 
probability, to /3-oxybutyric acid or to acetone bodies. The results 
of this computation are considered of sufficient importance to be 
included in table 38, which gives the total nitrogen for each day, the 
values for the normal excretion of carbon as computed with the ratio 
C : N = 0.79, the total carbon as actually determined, the difference 
which would be ascribed to /3-oxybutyric acid, and finally the values for 
/3-oxybutyric acid obtained by using 46.11 as the percentage of carbon 
in /3-oxybutyric acid. As the values for /3-oxybutyric acid actually 
determined are also given in this table, a comparison may readily be 
made. 

In general the determined values are somewhat lower than those 
obtained by computation. Occasionally, especially toward the end of 
the fast, the determined values are higher than the calculated values. 
On the whole, however, there is sufficient agreement to give an approxi- 
mate estimate as to the amount of /3-oxybutyric acid present. Indeed, 
taking everything into consideration and the regularity of the figures, 
it would appear that the computed values are probably more nearly 
accurate than those which were actually determined. In any event, 
the amounts here found are noticeably less than those computed from the 
laevo-rotation of the urine observed by Brugsch with Succi and by Grafe 
with his fasting woman. When the computed values are compared, it 
is found that the greatest excretion of /3-oxybutyric acid was from the 
sixteenth to the nineteenth days, when for 4 days an average amount 
of over 7 grams was excreted. 

The amounts of /8-oxybutyric acid present in these urines are not 
unlike those found in short fasting experiments or in experiments with 
a carbohydrate-free diet, and the stimulating effect of these acids upon 
metabolism is certainly not to be ignored. On the other hand, with the 
acid present in the body for so long a time, the subject might easily 
have become accustomed to its presence and therefore the reaction be 
less, as has been found in cases of severe diabetes. 

It is perfectly clear, however, that the amounts of /3-oxybutyric acid 
involved in these determinations are not sufficient to affect the respira- 



URINE. 285 

tory quotients to an appreciable degree, since, as Magnus-Levy 1 has 
pointed out, the excretion of 20 grams of /3-oxybutyric acid per day 
results in a lowering of the respiratory quotient only 0.006. The pres- 
ence of this acid is, however, sufficient to account for the increase in 
the ammonia excretion as the fast continued. The effort of the body 
to correct this undue acidity by combining ammonia with the acid is 
thus clearly shown. 

Bonniger and Mohr 2 found with Schenk much greater amounts of 
acid than were found or indeed calculated for the urine of L. Brugsch 3 
likewise found large amounts with Succi, and the excretion for Grafe's 4 
insane patient was also large. The uncertainty in the method for the 
quantitative determination of /3-oxybutyric acid does not permit use 
of these values for comparison, and it is sufficient to state that lsevo- 
rotatory /3-oxybutyric acid in appreciable amounts is excreted in the 
urine during fasting. 

Beginning on the afternoon of April 21, 1912, and continuing at 
least every other afternoon during the fast, a qualitative test for acetone 
in the breath was made, using the reagent of Scott- Wilson. 5 Several 
tests were also made each day after the fast, the latest being on May 
17 at 7 p. m. In every case the test showed acetone present, but it is 
impossible to draw quantitative deductions from the results. 

MINERAL METABOLISM. 

With normal man the mineral metabolism has two main paths for 
excretion — through the solid salts of the urine and through the feces. 
With L. the entire mineral excretion took place through the urine, if 
one excepts the small amount of sodium chloride excreted through the 
skin. Usually much stress is laid upon the determination of the acid 
radicles in the urine, namely, chlorine, sulphur trioxide, and phosphorus 
pentoxide, and but little attention is given to the calcium, magnesium, 
potassium, and sodium metabolism. The important relationship 
between phosphorus and calcium and the possible draft upon the 
skeletal tissue of the body made the determination of the mineral 
constituents of the urine desirable, and although a very large number of 
determinations were made of the various components of the urine of 
our subject, it was possible, by combining and apportioning the mate- 
rial, to secure a sufficient sample of urine for the determination of the 
mineral constituents. 

The analyses were made by Mr. J. C. Bock, who was at that time a 
member of the Laboratory staff and who had had previous experience 
in mineral analysis. Since special training was necessary for this inves- 

^agnus-Levy, Zeitschr. f. klin. Med., 1905. 56, p. 83. 

2 Bonniger and Mohr, Zeitschr. f. exp. Path. u. Therapie, 1906, 3, p. 675. 

3 Brugsch. Zeitschr. f. exp. Pathol, u. Therapie, 1905, 1, p. 419. 

4 Grafe, Zeitschr. f. physiol. Chemie, 1910, 65, p. 27. 

8 Scott-Wilson, Journ. Physiol., 1911, 42, p. 444. 



286 



A STUDY OF PROLONGED FASTING. 



tigation of the mineral metabolism during fasting, Mr. Bock was allowed 
the privilege, through the kindness of Dr. Rufus S. Cole of the Rocke- 
feller Hospital, of working with Dr. Francis H. McCrudden, at that 
time of the Rockefeller Hospital, and whose researches in mineral 
metabolism are too well known to need special mention here. Having 
acquired certain of Dr. McCrudden's methods and technique, Mr. 
Bock made a most careful analysis of the urine of L., determining the 
calcium, magnesium, sodium, and potassium, so that we have a fairly 

Table 39. — Mineral metabolism (urine excretion) in experiment with L. 



Date. 



Total. 




Calcium 
(Ca). 



Magnesium 
(Mg). 



I 
Aver- 
age per Total, 
day- 



Aver- 
age per 
day. 



Potassium 
(K). 



Total. 



Aver- 
age per 
day. 



Sodium 
(Na). 



Total. 



Aver- 
age per 
day. 



gm. 



gm. 



gm. 



gm. 



1.630 



368 
368 
368 
445 
445 
.883 
.883 
.883 
1.006 
1.006 



.070 
.779 

.552 

.463 

.199 



2.070 

.926 

.926 

.926 

I .276 

i .276 

f .154 

I .154 

I .154 

f .100 

I .100 



1.627 



.814 
.814 



.217 



.109 
.109 



i: 

{: 

: 

: 

: 

.; 
: 



676 
676 
644 
644 
643 
643 
787 
787 
656 
656 
585 
585 
606 
606 
430 
430 
.116 



.102 

.131 

.166 

.129 

.109 

.071 

.105 

.107 
.046 



.051 
.051 
.066 
.066 
.083 
.083 
.065 
.065 
.055 
.055 
.036 
.036 
.053 
.053 
.054 
.054 
.046 



determinations in urine for about 22 hours. 



URINE. 287 

complete picture of the mineral metabolism of this subject throughout 
the entire fast. As a rule, the urines for 2 days were combined, the 
determinations thus representing the amounts for 2-day periods. On 
two occasions it was necessary to combine the urines of 3 days, while 
the mineral metabolism was determined for the days preceding the 
fast on a sample representing the total urine for those 4 days. The 
results are given in table 39, the values per day for convenience being 
interpolated. 

Throughout the entire fast there was a material excretion of all four 
of these elements. The calcium excretion remained relatively constant, 
with a slight tendency to fall off as the fast progressed, particularly after 
the third week. The highest amount, 0.274 gram per day, was observed 
on the fifth and sixth days of fasting; the smallest amount, 0.131 gram, 
was found on the twenty-eighth and twenty-ninth days of the fast. 

With magnesium there was a very considerable increase in the first 
portion of the fast, and even for the last days of the fasting period the 
average was 0.052 gram, which was considerably more than the amount 
excreted per day in the four days prior to the fast, i. e., 0.034 gram. 

With the potassium there was a very great decrease as the fast pro- 
gressed. The largest amount, 1.63 grams, was found on the first day 
of the fast, the excretion steadily falling until on the twenty-eighth and 
twenty-ninth days of the fast it reached 0.585 gram. It is interesting 
to note that at all times there was measurably over 0.5 gram of potas- 
sium excreted per day. 

With sodium we find perhaps greater variations than for any other 
of these four substances, there being a rapid decrease for the first few 
days, followed by a more moderate but steady decrease for the remain- 
der of the fast. The values range from 2.07 grams on the first day of 
the fast to 0.036 gram on the twenty-eighth and twenty-ninth days. 
It is thus seen that the decreases in the minerals excreted were by no 
means parallel. 

Relationships of the Mineral Constituents. 

In studying the mineral metabolism of the subject L. as indicated 
by the results of the urinary analyses given in table 39 we see in- 
stantly the great advantage in having the subject drink only distilled 
water, for the entire mineral output is then derived solely from the 
body tissue; we can therefore consider the relationships between 
calcium and magnesium and between sodium and potassium without 
the discussion being complicated by the possibility of varying amounts 
of these elements in the drinking-water. 

While a continuous decrease is evident in the excretion of all four 
elements, it is likewise clear that the diminution in the excretion is 
more marked with sodium than with any other element. Fortunately 
the relations between potassium and sodium and calcium and magne- 



288 A STUDY OF PROLONGED FASTING. 

shim in the animal body have been recently carefully studied, 1 and we 
may more advantageously study the excretion of these four elements by 
noting the ratios of their oxides to each other than in any other way. 
Consequently the percentages of calcium oxide and magnesium oxide 
excreted have been computed, using the total of the calcium oxide and 
magnesium oxide as 100. Similarly the potassium and sodium oxides 
have been added together and the percentages of potassium oxide and 
sodium oxide computed. These percentages, which are given in table 
40, were computed for each sample of urine analyzed. 

Prior to the fast there was about eight times as much calcium oxide 
excreted in the urine as magnesium oxide, but this relationship became 
4 to 1 on the first fasting day and subsequently the percentages were 
fairly constant at approximately 75 per cent for the calcium oxide and 
25 per cent for the magnesium oxide. It was not until the third day 
with food following the fast that this relationship was disturbed, when 
the original percentages prior to the fast were again approximated. 
According to these ratios, then, there were relatively much larger 
amounts of magnesium excreted during fasting than on the food days. 

In considering the ratios for the potassium oxide and sodium oxide, 
we find that the problem is considerably complicated by the fact that 
a certain amount of sodium chloride exists in the body which is very 
loosely, if at all, combined in the tissues. Consequently, on the first 
day of the fast, only 41.3 per cent of the total alkali was found in the 
form of potassium oxide and 58.7 per cent in the form of sodium oxide. 
On the next day the proportion was reversed, and from that time until 
the eighteenth day there was a tendency toward a gradually increasing 
percentage of potassium oxide. For the remaining days of the fast, 
about 90 per cent of the alkali was in the form of potassium oxide and 
10 per cent or less in the form of sodium oxide. One specimen of urine — 
that for the twenty-eighth and twenty-ninth days — shows the high 
percentage of 93.6 for potassium oxide as compared with 6.4 per cent 
for sodium oxide. 

It is obvious from the values for the potassium and sodium that 
shortly after the excretion of the uncombined sodium chloride on the 
first few days, the body excreted the potassium and sodium from muscle 
substance and we then have the large differences between these two 
elements; in the last part of the fast, approximately nine or ten times 
more potassium was excreted than sodium. This far exceeded the 
ordinary proportion between potassium and sodium in muscle given 
as a result of the analyses of Bunge. 2 

The relationship between calcium and magnesium in animal tissues 
has been extensively studied, particularly with dogs and horses. 

Gerard, Ann. de l'Inst. Pasteur, 1912, 26, p. 12. 

s See Aron, Oppenheimer's Handbuch der Biochemie des Menschen u. der Tiere, 1909, 1, p. 84, 
where the relationship is given as 5 to 6 times as much potassium as sodium. 



URINE. 289 

Table 40. — Distribution of mineral metabolism (urine excretion) in experiment with L. 





Day 
of fast. 


Calcium and magnesium. 


Potassium and sodium. 




Total, 
(grams) 


CaO. 


MgO. 


Total. 

(grams) 


K 2 0. 


Na 2 0. 


Date. 


Grams. 


Per cent. 
/bX100\ 


Grams. 


Per cent. 
/dX100\ 


Grams. 


Per cent. 


Grams. 


Per cent. 
(xXJOO) 






A 


B 


c 


D 


£ 


F 


G 


H 


I 


J 


1912. 
\pr. 10-11 




1 




















11-12 




>1.942 

0.381 
[l.545 


1.718 

0.304 
1.020 


88.5 

79.8 
66.0 


0.224 

0.077 
0.525 


11.5 

20.2 
34.0 












12-13 


4.752 

8.688 


1.964 
4.944 


41.3 
56.9 


2.788 
3.744 


58.7 
43.1 


13-14 




14-15 

15-16 

16-17 

17-18 


1st . 
2d.. 
3d.. 

4th. 


18-19...... 

19-20 


5th. 
6th. 


|l.090 
il.409 


0.765 


70.2 


0.325 


29.8 


4.224 


3.480 


82.4 


0.744 


17.6 


20-21 

21-22 

22-23 


7th. 
8th. 
9th. 


1.062 


75.4 


0.347 


24.6 


3.816 


3.192 


83.6 


0.624 


16.4 


23-24 

24-25 

25-26 

26-27 


10th. 
11th. 
12th. 
13th. 


)o.854 
Jo. 819 


0.616 
0.604 


72.1 
73.7 


0.238 
0.215 


27.9 
26.3 


2.692 


2.424 


90.0 


0.268 


10.0 












27-28 

28-29 


14th. 
15th. 


jo. 893 
]o.856 


0.659 


73.8 


0.234 


26.2 


2.253 


1.960 


87.0 


0.293 


13.0 


29-30 

\.pr. 30-May 1.. 


16th. 
17th. 


0.597 


69.7 


0.259 


30.3 






















Vlay 1-2 

2-3 


18th. 
19th. 


Jo. 897 


0.701 


78.1 


0.196 


21.9 


1.766 


1.628 


92.2 


0.138 


7.8 


3-4 

4-5 


20th. 
21st.. 


Jo. 838 


0.664 


79.2 


0.174 


20.8 


1.728 


1.552 


89.8 


0.176 


10.2 


5-6 

6-7 


22d.. 
23d.. 


Jo. 664 


0.500 


75.3 


0.164 


24.7 


1.772 


1.548 


87.4 


0.224 


12.6 


7- 8 

8- 9 


24th. 
25th. 


Jo. 652 


0.468 


71.8 


0.184 


28.2 


2.070 


1.896 


91.6 


0.174 


8.4 


9-10 
10-11 


26th. 
27th. 


Jo. 596 


0.428 


71.8 


0.168 


28.2 


1.726 


1.580 


91.5 


0.146 


8.5 


11-12 

12-13 


28th. 
29th. 


Jo. 520 


0.366 


70.4 


0.154 


29.6 


1.504 


1.408 


93.6 


0.096 


6.4 


13-14 

14-15 

15-16 


30th. 
31st. . 


Jo. 558 

Jo. 263 
0.147 


0.385 


69.0 


0.173 


31.0 


1.602 


1.460 


91.1 


0.142 


8.9 


16-17 




0.202 
0.134 


76.8 
91.2 


0.061 
0.013 


23.2 

8.8 


1.179 
0.203 


1.035 
0.140 


87.8 
69.0 


0.144 
0.063 


12.2 
31.0 


17-18* 









1 Determinations in urine for about 22 hours. 

Toyonaga 1 in Tokio found that in the muscles there is always less 
calcium than magnesium, thus confirming the analyses of Katz, 2 but 
in the glands there is always more calcium than magnesium. 

Aloy, 3 studying the calcium and magnesium content of muscle, found 
about twice as much magnesium as calcium. Perhaps of more interest 

toyonaga, Bui. Coll. of Agr., Tokio, 1902-1903, 5, p. 143, and p. 455; also 1904-1905, 6, pp. 
89 and 357. 

2 Katz, Archiv f. d. ges. Physiol., 1896, 63, p. 1. 
3 Aloy, Compt. rend. Soc. Biol., 1902, 54, p. 604. 



290 A STUDY OF PROLONGED FASTING. 

in this particular study are the observations of Magnus-Levy, 1 who ana- 
lyzed the body of a man who had committed suicide. His results 
showed about three times as much magnesium as calcium in the muscles. 
On the basis of all the analyses prior to those of Magnus-Levy, Aron 2 
gives the average figure for the relationship of magnesium to calcium 
in the muscle of dogs as 1 to 0.54-0.60 and in the muscle of horses as 
1 to 0.34. In general, then, we may assume that there is approxi- 
mately three times as much magnesium as calcium in the muscle of man. 

We may consider the magnesium excretion as more nearly an index 
of the muscle disintegration than calcium, for while there is, to be sure, 
a small percentage of magnesium oxide in bone, there is a much larger 
available supply of calcium oxide in the form of bone which is unques- 
tionably drawn upon. The calcium-oxide excretion is therefore the 
resultant of two factors, i. e., muscle disintegration and bone disin- 
tegration, while magnesium oxide is derived almost exclusively from 
the non-osseous tissue. 

It is a matter of regret that the small amount of magnesium present 
in the urine and the necessity for combining the samples for several 
days renders it very difficult to make an exact comparison between 
the magnesium excretion and the other urinary constituents. In 
general, however, the magnesium excretion follows approximately 
the nitrogen excretion. Of striking interest is the fact that, on the 
first day of the fast, an extraordinarily low amount of both magnesium 
and calcium were excreted. The increment in the magnesium excre- 
tion on the second day (i. e., 100 per cent) was not approximated by 
the excretion of any other element in the body. 

The results of the analyses reported by Cathcart for Beaute, while 
relatively few in number, are in full conformity with our findings, 
save that the magnesium excretion on the days prior to the fast is 
much larger than that found with L. Furthermore, L. excreted con- 
siderably more calcium per day than did Beaute" during the fasting 
period. 

In the food period following the fast, it is interesting to note the 
striking fall in the excretion of all of the minerals save sodium. For 
the first 2 food days there was but half as much calcium excreted as 
on the last day of the fast, about one-third as much magnesium, seven- 
tenths as much potassium, and about the same amount of sodium. 
On the last day of the food period after the fast, the calcium was seven- 
tenths that of the last fasting day, the magnesium one-seventh, the 
potassium one-fifth, and the sodium about the same as on the last fast- 
ing day. 

The amounts of calcium, magnesium, potassium, and sodium intake 
are unknown for these days, yet these elements must have been present 
in the food taken. It is obvious that the effect of the ingestion of a large 

Magnus-Levy, Biochem. Zeitschr., 1910, 24, p. 363. 

2 Aron, Oppenheimer's Handbuch der Biochemie des Menschen u. der Tiere, 1909, 1, p. 88. 



URINE. 291 

amount of carbohydrates upon the mineral metabolism was consider- 
able, resulting in a marked retention of the inorganic salts introduced, 
with a noticeable lessening in the attack upon the storage of mineral 
matter in the body. On the other hand, it should be remembered that 
on these days fecal matter was passed in considerable amounts and we 
may have here to deal only with the disturbance in the paths of excre- 
tion of mineral matter. Thus, in the total amount of fecal material 
excreted between 5 p. m. and 8 a. m. on the first food day following the 
fast, there were excreted 1.78 grams of calcium oxide and 0.748 gram 
of magnesium oxide, as determined by Mr. Bock on the dry matter of 
feces. On this basis 30 per cent of the earthy alkali was magnesium. 

REDUCING POWER. 

The presence, even in fasting urines, of reducing substances other 
than dextrose, has frequently been noted. Munk 1 especially has 
studied this subject and made extensive observations of the reducing 
power of the urines in Breithaupt's experiment, using a reduction 
method developed by himself 2 and further elaborated and tested by 
Hagemann. 3 Munk's method gives as the reducing power for normal 
urines from 0.16 to 0.47 per cent, with an average of 0.3 per cent. In 
the fasting experiment with Breithaupt he found in the urine of the 
last 2 food days as high as 7.7 grams of reducing substance (calculated 
as dextrose). Even in fasting periods amounts were obtained ranging 
from 3 to 7 grams per day. 4 Furthermore, he noted very considerable 
fluctuations from day to day. Munk considers the reducing action to 
be due in large part to the formation of glykuronic acid. The reducing 
power of the urine bore no relationship to the amount of carbohydrate 
ingested, but there was a tendency to parallel the protein disintegrated. 
In connection with the experiment with L., Dr. A. W. Peters suggested 
that it would be desirable to test the reducing power of the urine. As 
Dr. Peters had previously developed an accurate method 5 for testing 
the amounts of reducing substances in urine, this could be done to 
advantage, and accordingly the determinations were made for each day 
of the fast by W. F. O'Hara under Dr. Peters's supervision. The results, 
expressed in terms of dextrose, are given in table 41. 

The Peters method gives considerably less reducing substance in the 
urine than the method of Munk. Thus, for normal urines, Peters has 
found in this laboratory from 0.03 to 0.12 per cent as compared with 
the values of 0.16 to 0.47 per cent found by Munk. Since the observa- 
tions of the urine in our fasting experiment were to be wholly compara- 
tive, either method was suitable for studying the variations in the 

^unk, Archiv f. path. Anat. u. Physiol., 1893, 131, Supp., p. 138. 
2 Munk, Archiv f. path. Anat. u. Physiol., 1886, 105, p. 73. 
3 Hagemann, Archiv f. d. ges. Physiol., 1888, 43, p. 501. 
4 Munk, Archiv f. p*th. Anat. u. Physiol., 1893, 131, Supp., p. 68, table 7. 
6 See description of method by Dr. Peters in Benedict and Joslin, Carnegie Inst. Wash. Pub. 
176, 1912, p. 8. 



292 



A STUDY OF PROLONGED FASTING. 



reducing power as the fast progressed, regardless of any inherent differ- 
ences which might exist in the two methods. 

The reducing power of the fasting urines in the experiment with L. 
was at all times well within normal limits, averaging not far from the 
values observed on control normals in this laboratory. The largest 
amount on the fasting days was 498 milligrams for the fourth and six- 
teenth days, and the smallest amount was 296 milligrams for the third 
and sixth days. 

A large amount of reducing power was found on the first day of food 
following the fast. On this day the subject ate in a relatively short 
time about 500 grams of carbohydrate, chiefly in the form of soluble 



Table 41.- 


— Total reducing power 


of urine in experiment with L. 


Date. 


Day of 

fast. 


Reducing 
power (as 
dextrose). 


Date. 


Day of 

fast. 


Reducing 
power (as 
dextrose) . 


1912. 
Apr. 14-15 

15-16 

16-17 

17-18 

18-19 

19-20 

20-21 

21-22 

22-23 

23-24 

24-25 

25-26 

26-27 

27-28 

28-29 

29-30 

Apr. 30-May 1 . . . 


1st 

2d 

3d 

4th ... . 

5th.... 

6th.... 

7th.... 

8th.... 

9th.... 
10th.... 

11th 

12th 

13th 

14th.... 
15th.... 
16th.... 
17th.... 


mg. 
450 
414 
296 
498 
328 
296 
376 
376 
414 
356 
342 
396 
343 
384 
328 
498 
434 


1912. 
May 1- 2. . . 

2- 3 . . . 

3- 4 . . . 

4- 5 . . . 

5- 6 . . . 
6-7... 
7-8... 
8-9... 
9-10... 

10-11.. . 
11-12... 
12-13 . . . 
13-14... 
14-15. .. 
15-16 . . . 


18th . . . 
19th . . . 
20th... 
21st. . . 
22d 
23d 

24th . . . 
25th... 
26th . . . 
27th . . . 
28th. . . 
29th... 
30th . . . 
31st. . . 


mg. 
396 
376 
376 
312 
342 
328 
376 
356 
376 
328 
342 
376 
396 
342 
4441 
267 
x 246 


16-17. . . 




17-18 









determined in urine for about 22 hours. 

dextrose, the diet consisting of honey and the juice of grapes, oranges, 
and lemons. Unquestionably this amount exceeded his carbohydrate 
tolerance on that day and 4.44 grams of dextrose were therefore 
excreted in the urine. It is possible that we find here with man the 
condition with dogs described by Hofmeister 1 as "hunger diabetes," 
and it may be an interesting confirmation of his theory. 

Recently, also, Rietschel 2 has noted that the fasting of infants has 
resulted in a marked lowering of their tolerance for carbohydrate sub- 
stances. This lowering has been so noticeable as to lead Rietschel to 
warn clinicians in the following words against the undue use of starving 
in treating pathological cases: 

"Dass der Hunger spez. die absolute Nahrungsentziehung fiirdengesunden, 
wie besonders fur den ernahrungsgestorten Saugling auch schwere Gefahren 
nach sich Ziehen kann, ist heute allgemein anerkannt." 

hofmeister, Archiv f. exp. Path. u. Pharm., 1889-1890, 26, p. 355. 
2 Rietschel, Heubner's Festschrift, Berlin, 1913, p. 516. 



URINE. 293 

To attempt an analysis of the reducing substances in the urine during 
the fasting period would be somewhat difficult. A certain portion 
has already been ascribed to creatinine and to uric acid, but there is no 
definite relationship between the amounts of uric acid, creatinine, and 
the reducing power to be noted in the results obtained in this fasting 
experiment. 

CARBON IN URINE. 

While the excretion of carbon in the form of carbon dioxide is of great 
significance as indicating the total amount of energy transformed into 
heat in the body, nevertheless when a study of the total loss of body 
material is of importance, as it is during complete fasting, the carbon 
in the urine must be taken into consideration. Since we were also 
making a study of the energy output of the urine by burning samples 
in a calorimetric bomb, it was relatively simple to combine the deter- 
minations of the energy output and the carbon content. Conse- 
quently, after the dried urine had been burned inside the bomb, the 
carbon dioxide produced in the combustion was allowed to escape from 
the vessel through weighed soda-lime tubes, in accordance with the 
method of Fries. 1 These determinations were skilfully carried out by 
Mr. A. W. Cornell, of the Laboratory staff. 

The description of the method of preparation and drying of the urine 
samples for the bomb calorimeter has already been given in the dis- 
cussion of the total solids in the urine. 2 From the weight of the carbon 
dioxide in the soda-lime tubes and the weight of the urine, the amount 
of carbon per day excreted in the urine is readily computed. These 
values are given in table 42. 

The determination of carbon in urine has been for many years a 
subject of research in this laboratory and in the chemical laboratory j 
of Wesleyan University. Various methods have been tried, including 
the moist combustion process, drying with and without the addition 
of salicylic acid, and with and without the use of the cellulose filter 
blocks recommended by Kellner. 3 The method which gives the largest 
percentage of carbon is presumably the best one, and this has been our 
criterion. No method that we have thus far used approaches the large 
percentage of carbon which is obtained by the method previously de- 
scribed, 4 namely, drying first with 50 milligrams of salicylic acid, then 
transferring to a nickel capsule, drying in a desiccator until ready to burn, 
and finally burning in compressed oxygen in a bomb calorimeter, and 
allowing the carbon dioxide to escape into soda-lime. When the heat 
of combustion is desired, the determination of carbon occupies but 
a few moments additional, thus providing the simplest and best method 
for obtaining the required values. The preliminary operations of drying 
require but little attention from the assistant. 

^ries, Journ. Am. Chem. Soc, 1909, 31, p. 272. 

2 See p. 243. 

3 Kellner, Landw. Jahrb., 1896, 47, p. 297. 

4 Higgins and Benedict, Am. Journ. Physiol, 1911, 28, p. 291. 



294 



A STUDY OF PROLONGED FASTING. 



The total amount of carbon excreted in the urine ranged, according 
to the values in table 42, from 5.82 grams on the first day of the fast 
to 11.88 grams on the fourth day of the fast. In the first half of the 
fasting period the carbon excretion averaged somewhat above 10 grams 
per day, but in the latter part the excretion was not far from 8 grams. 



Table 42. — Nitrogen, carbon, and energy of urine in experiment with L. 


Date. 


Day of 
fast. 


Nitrogen. 


Carbon. 


Carbon 
per gram 
of nitro- 


Energy of urine. 




Per gram of 


Per gram 










(C:N) 


Total. 


nitrogen. 


of carbon. 












(Cals. : N) 


(Cals. : C) 


1912. 




gm. 


gm. 


gm. 


cals. 


cals. 


cals. 


Apr. 12-13 




14.48 


11.41 


0.788 


129 


8.91 


11.30 


13-14 




11.54 


9.08 


.787 


104 


9.01 


11.45 


14-15 


1st 


7.10 


5.82 


.820 


65 


9.15 


11.17 


15-16 


2d 


8.40 


7.99 


.951 


89 


10.60 


11.14 


16-17 


3d 


11.34 


10.35 


.913 


118 


10.40 


11.40 


17-18 


4th.... 


11.87 


11.88 


1.001 


134 


11.29 


11.28 


18-19 


5th 


10.41 


10.69 


1.027 


123 


11.82 


11.51 


19-20 


6th 


10.18 


10.42 


1.024 


116 


11.40 


11.13 


20-21 


7th.... 


9.79 


9.06 


.925 


104 


10.62 


11.48 


21-22 


8th 


10.27 


10.30 


1.003 


116 


11.30 


11.26 


22-23 


9th ... . 


10.74 


10.92 


1.017 


124 


11.54 


11.36 


23-24 


10th 


10.05 


9.92 


.987 


111 


11.04 


11.19 


24-25 


11th.... 


10.25 


9.59 


.936 


110 


10.73 


11.47 


25-26 


12th 


10.13 


9.05 


.893 


105 


10.36 


11.60 


26-27 


13th 


10.35 


10.15 


.981 


114 


11.01 


11.23 


27-28 


14th 


10.43 


9.95 


.954 


111 


10.64 


11.16 


28-29 


15th.... 


8.46 


8.71 


1.030 


95 


11.23 


10.91 


29-30 


16th 


9.58 


11.39 


1.189 


123 


12.84 


10.80 


Apr. 30-May 1 . 


17th.... 


8.81 


10.91 


1.238 


117 


13.28 


10.72 


May 1-2 


18th 


8.27 


9.65 


1.167 


104 


12.58 


10.78 


2-3 


19th 


8.37 


9.56 


1.142 


105 


12.54 


10.98 


3- 4 


20th.... 


7.69 


8.07 


1.049 


91 


11.83 


11.28 


4-5 


21st 


7.93 


8.59 


1.083 


95 


11.98 


11.06 


5- 6 


22d 


7.75 


8.40 


1.084 


93 


12.00 


11.07 


6- 7 


23d 


7.31 


7.25 


.992 


88 


12.04 


12.14 


7-8 


24th.... 


8.15 


8.68 


1.065 


95 


11.66 


10.94 


8-9 


25th.... 


7.81 


8.68 


1.099 


91 


11.65 


10.61 


9-10 


26th 


7.88 


8.56 


1.086 


90 


11.42 


10.51 


10-11 


27th 


8.07 


8.23 


1.020 


90 


11.15 


10.94 


11-12 


28th 


7.62 


7.73 


1.014 


85 


11.15 


11.00 


12-13 


29th 


7.54 


7.94 


1.053 


87 


11.54 


10.96 


13-14 


30th.... 


7.83 


7.95 


1.015 


87 


11.11 


10.94 


14-15 


31st 


6.94 


7.37 


1.062 


80 


11.53 


10.85 


15-16 




4.83 
3.81 


7.13 

4.28 


1.476 
1.123 


74 
45 


15.32 
11.81 


10.38 
10.51 


16-17 









In the fasting experiments at Wesleyan University the carbon 
content of the urine increased noticeably on a number of days as the 
fast continued. This increase in carbon was accompanied by an 
increase in the energy content, which was attributed at the time to the 
presence of a large amount of /3-oxybutyric acid. A similar increment 
in the carbon content of fasting urine was noted with an insane patient 
by Benedict and Diefendorf. 1 This excess of carbonaceous material 

Benedict and Diefendorf, Am. Journ. Physiol., 1907, 18, p. 362. 



URINE. 295 

became apparent when the ratio between the carbon and nitrogen 
of normal urine was compared with that obtained in these fasting 
experiments. 

Cakbon-Nitrogen Ratio. 

Since the carbon content of the urine naturally fluctuates to a certain 
degree with the nitrogen content, it is obvious that the determinations 
of the carbon alone would not have the significance of the ratio between 
the carbon and nitrogen, for we are interested not so much in the carbon 
normally accompanying nitrogen in the urine as in the carbon other 
than that in nitrogenous material. With an ordinary diet, essentially 
constant ratios have been found for normal urines by various investi- 
gators, averaging not far from 0.8 gram of carbon for each gram of 
nitrogen excreted in the urine. Thus Benedict and Milner 1 found an 
average carbon-nitrogen ratio of 0.73 for 58 metabolism experiments 
upon normal individuals with rest and work in the respiration calori- 
meter at Wesleyan University. The variations in these experiments 
were, in general, very small, the carbon-nitrogen ratio ranging from 
0.67 to 0.89. Richardson, 2 working with fewer urines, obtained a 
carbon-nitrogen ratio varying from 0.74 to 1.01, with an average of 0.88. 
Magnus- Alsleben 3 concludes that with healthy individuals the carbon- 
nitrogen ratio will not pass beyond the limits of 0.7 and 1.0, regardless 
of diet. Loewy, 4 Pregl, 6 Reale, 6 and others reported values with normal 
individuals essentially within these limits. 

The composition of the diet seems to have but little effect upon this 
ratio. Benedict and Milner noted no appreciable variation due to 
change in diet containing a preponderance of either carbohydrate or 
fat, but Tangl 7 found an average ratio of 0.96 on days when the diet 
was rich in carbohydrates and poor in fat, which was considerably 
higher than the average carbon-nitrogen ratio of 0.75 which was found 
when the diet was poor in carbohydrate and rich in fat. Notwith- 
standing the fact that the diet on these days was extraordinary, it may 
be noted that Tangl's figures fell within the limits set by Magnus- 
Alsleben. 

Moderate muscular work has been shown both by Tangl and by 
Benedict and Milner to increase the ratios little, if any, over those for 
rest, although the severe muscular exercise incidental to the strenuous 
work of a Marathon race was found by Higgins and Benedict 8 to result 
in distinctly abnormal carbon-nitrogen ratios in a number of cases, a 
ratio as high as 1.517 being found in one instance. 

Benedict and Milner, U. S. Dept. Agr., Office Expt. Sta. Bui. 175, 1907, p. 145. 
2 Richardson, Bulletin Mt. Hope Retreat Laboratory, 1900. Cited in Maly's Jahrsb. d. Tier- 
Chemie, 1901, 31, p. 703. 

3 Magnus-Alsleben, Zeitschr. f. klin. Med., 1909, 68, p. 358. 

4 Loewy, Verhndl. der physiol. Gesellsch. zu Berlin, 1905-1906, p. 11. 

6 Pregl, Archiv f. d. ges. Physiol., 1899, 75, p. 87. 

6 Reale, Biochem. Zeitschr., 1912, 47, p. 355. 

'Tangl, Archiv f. Anat. u. Physiol., 1899, Physiol. Abth. Supp., p. 251. 

8 Higgins and Benedict, Am. Journ. Physiol., 1911, 28, p. 291. 



296 A STUDY OF PROLONGED FASTING. 

With normal urines, therefore, one may conclude that the carbon- 
nitrogen ratio may vary from 0.67 to 1.0. Many interesting cases 
are recorded in which the disturbance of the carbon-nitrogen ratio has 
been found. In the fasting experiments at Wesleyan University 
carbon-nitrogen ratios during fasting ranged from a minimum of 0.660 
to a maximum of 1.293. On 7 days out of 43 the ratio was over 1, 
these 7 days being the last 4 days of a 7-day fast and the last 3 of a 
4-day fast. In certain types of fever, also, Magnus-Alsleben 1 found 
an increase of the carbon-nitrogen ratio, while in others there was a 
marked decrease. He also reports three cases in which extremely 
high ratios were obtained after severe muscular work, these ratios 
being 3.262, 1.926, and 1.038 respectively. It is thus clear that under 
conditions which result in an abnormal katabolism, disturbances in 
the carbon-nitrogen ratio are found, and conversely a disturbance of 
this ratio may be taken as prima facie evidence of a distinctly disturbed 
katabolism. 

The carbon-nitrogen ratio has been computed for each day of the 
fasting experiment with L. and likewise for the food days prior and sub- 
sequent to the fast. These values are included in table 42, together 
with the values for the total nitrogen. On the 2 days before the fast 
the ratio was very constant, averaging 0.79. It then rose rapidly until 
the fourth day, when it was slightly over 1.0 and remained at approxi- 
mately 1 until the maximum level was reached between the sixteenth 
and nineteenth days of fasting. The very high value of 1.476 on the 
first day with food after the fast is in part explained by the excretion 
of 4.44 grams of dextrose in the urine. It is seen from these ratios, 
therefore, that the urine excretion after the first day or two regularly 
contained some nitrogen-poor and carbon-rich substance which, from 
all evidence, appears to be /3-oxybutyric acid. Since the 2 days with 
food before the fast agree so perfectly, we have felt justified in using the 
average value of 0.79 for computing the normal amount of carbon 
accompanying nitrogen in the indirect computation of the amount of 
jS-oxybutyric acid present in the urine. (See column B,table 38, page283.) 

The values found with L. during fasting are materially higher 
throughout the entire fast than those reported by Munk for Breithaupt, 
for on 6 days of the fast Munk found no difference in the carbon- 
nitrogen ratio between the fasting days and the 2 days with food 
following the fast. On the last 6 days of a 3-weeks fast Grafe 2 found 
extraordinarily high carbon-nitrogen ratios as follows: 1.714, 1.642, 
2.016, 1.873, 1.63, 1.53. On the first food day the ratio fell to 0.746. 
The three observations of Pettenkofer and Voit 3 may also be cited, 
these investigators finding on the first fasting day an average of 0.7 
as the carbon-nitrogen ratio. 

1 Magnus-Alsleben, Zeitschr. f. klin. Med., 1909, 68, p. 358. 
2 Grafe, Zeitschr. f. Physiol. Chem., 1910, 65, p. 21. 
3 Pettenkofer and Voit, Zeitschr. f. Biol., 1866, 2, p. 459. 



URINE. 297 

ENERGY OF URINE. 

In studying the metabolism of a fasting man, although we are par- 
ticularly interested in the energy transformed in the body and leaving 
the body as heat, a complete picture of the total breaking-down of 
tissue and loss of body material can not be had without a knowledge of 
the potential energy of unoxidized material in the urine throughout 
the fasting period. Determinations of the heat of combustion were 
made by Mr. A. W. Cornell, the results given in table 42 being always 
the average of two or three well-agreeing analyses. 

During the fasting period the total amount of energy lost in the 
urine ranged from 65 calories on the first day to 134 calories on the 
fourth day. There was a general tendency after the fourth day for 
the values to fall off gradually as the fast continued; excluding the first 
day, the smallest amount (80 calories) was found on the last day of the 
fast. The energy was also determined for the 2 days immediately 
preceding the fast, the values being 129 and 104 calories respectively. 
On the days with food following the fast, very small amounts of energy 
were found, these being 74 calories on the first and 45 calories on the 
second food day. 

Calorie-Nitrogen Ratio. 

Since the total amount of energy lost per day may vary with the 
amount of nitrogen excreted, and since there will always be a certain 
amount of potential energy normally accompanying each gram of 
nitrogen in the urine, it is important to compute the number of calories 
per gram of nitrogen. The results of such computation are also given 
in table 42. 

On the 2 days before the fast, there were about 9 calories daily per 
gram of nitrogen. The ratio remained practically unchanged on the 
first day of the fast, but for the next 4 days it showed a distinct ten- 
dency to increase. The highest ratio found during the fasting period 
was 13.28 calories per gram of nitrogen on the seventeenth day, and the 
lowest was 9. 15 calories on the first day of the fast. It is thus seen that, 
except on the first day, the ratios throughout the fast were compara- 
tively high, exceeding those found for the preceding food days. The 
ratios for the 2 days with food following the fast are influenced by the 
fact that on the first day, when the diet contained an excessive amount 
of carbohydrate, there was a measurable amount of sugar excreted 
(4.44 grams), which would obviously increase the energy but have no 
effect upon the nitrogen; furthermore, on the second day a very small 
amount of nitrogen was excreted. Neither of these values can of course 
be looked upon as obtained under normal conditions. 

With normal subjects and an ordinary diet, the ratio of calories 
to nitrogen is essentially constant. Thus Benedict and Milner 1 found 

Benedict and Milner, U. S. Dept. Agr., Office Exp. Sta. Bull. 175, 1907, p. 145. 



298 A STUDY OF PROLONGED FASTING. 

an average calorie-nitrogen ratio of 8.09 for 58 rest and work experi- 
ments with normal individuals in the respiration calorimeter at Wes- 
leyan University, the variations being extremely small, ranging from 
7.3 to 8.94. When unbalanced diets are taken, this ratio may be some- 
what altered. Tangl 1 reports the ratio considerably higher on the 
days when the diet was rich in carbohydrates and poor in fat, the ratio 
becoming as high as 11.67 on the carbohydrate-rich days and falling 
to 9.63 on the carbohydrate-poor days. On the other hand, Benedict 
and Milner noted no appreciable change due to diet. 

In the earlier fasting experiments the energy in the urine has been 
rarely determined, the most extensive investigation being that in Wes- 
leyan University; 2 in one experiment a calorie-nitrogen ratio was found 
ranging from 8.0 to 19.75. 

The large ratios found during fasting experiments are unquestionably 
to be explained by an excretion of nitrogen-poor, carbon-rich material, 
which is chiefly /3-oxybutyric acid. Unfortunately, in the earlier 
observation of Benedict and Diefendorf, in which the very high ratio 
of 19.75 was found, direct evidence of the excretion of j8-oxybutyric 
acid was not obtained, as the determinations were not then feasible. 
Assuming that the high calorie-nitrogen ratio is due to the presence of 
acetone bodies, it can be seen that during the fasting experiment of 
the subject L. the highest acidosis as measured by this means occurred 
between the fifteenth and twenty-fifth days, high ratios prevailing for 
this entire period. 

Calorie-Carbon Ratio. 

While it is perfectly possible to have a carbon-rich, nitrogen-free 
substance in the urine which would profoundly affect the calorie- 
nitrogen ratio, the presence of carbonaceous material in all energy- 
producing material found in urine would lead one to suppose that the 
relationship between calories and carbon would be much more regular 
than that between the calories and the nitrogen. The calorie-carbon 
ratios have been computed for this experiment and are included in 
table 42. The striking irregularities in the other ratios given in this 
table are entirely absent in the calorie-carbon ratio, for they remain 
remarkably constant under all conditions. The values for the entire 
series, including both the first and second food periods, range only 
between 12.14 on the twenty-third day of the fast and 10.38 on the first 
day with food after the fast. In the fasting period itself the minimum 
value was 10.51 on the twenty-sixth day. It is thus seen that during 
the total 31 days of the fast there were, on the average, 11.12 calories 
for each gram of carbon. 

The average calorie-carbon ratio in these fasting urines, namely, 
11.12 calories, is almost exactly the same as that found by Higgins and 

^angl, Archiv f. Anat. u. Physiol., 1899, Physiol. Abth. Supp., p. 251. 

2 Benedict, Carnegie Inst. Wash. Pub. 77, 1907, p. 493, and Benedict and Diefendorf, Am 
Journ. Physiol., 1907, 18, p. 362. 



URINE. 299 

Benedict, 1 their average calorie-carbon ratio for 18 specimens of urine 
excreted after severe muscular exercise incidental to a Marathon race 
being 11.02. This is also very close to the average calorie-carbon ratio 
obtained by Benedict and Milner, namely, 10.96. The striking uni- 
formity in the ratio between calories and carbon again emphasizes the 
importance of the development of some simple, rapid method of 
determining the carbon in urine which will not require the employment 
of a complicated bomb calorimeter. 

Still another relationship may be studied by comparing the total 
potential energy in the urine with the total estimated heat output. 
This comparison, however, is made in another section of the report. 
(See table 64, page 414.) The values given in this table show clearly 
that the total amount of energy excreted in the urine by the fasting 
subject L. equals approximately 8 to 10 per cent of the daily quota, 
and hence may not be neglected in any consideration of the energy 
lost by this man as the fast continued. 

Wiggins and Benedict, Am. Journ. Physiol., 1911, 28, p. 291. 



MICROSCOPY OF URINE AND TESTS FOR ALBUMIN. 
By Harry W. Goodall, M. D. 

The heat test was used in making the albumin determinations. For 
the sake of uniformity the microscopic examination was made as fol- 
lows : Two 15 mm. X 15 mm. cover-glass fields were examined with each 
specimen, 20 minutes being given to searching for and counting casts. 
The urine was centrifuged at a uniform rate for 5 minutes. The results 
of the tests are given herewith. 

DETAILED RESULTS. 

April llf-15 (first day of fast). — Albumin, absent. Sediment, no casts, 
blood, or pus; rare round cell, occasional squamous cell; little mucus. 

April 15-16 (second day of fast). — Albumin, absent. Sediment, no casts 
or pus; rare normal red blood corpuscle, rare small round cell, 
occasional squamous cell; little mucus. 

April 16-17 (third day of fast). — Albumin, absent. Sediment, one hyaline 
cast, small diameter, no blood or pus; few small round cells, rare large 
caudate cell, numerous squamous cells; rare spermatozoa, normal 
in appearance; little mucus. 

April 17-18 (fourth day of fast). — Albumin, absent. Sediment, no casts, 
blood, or pus; few small round and squamous cells, little mucus. 

April 18-19 (fifth day of fast). — Albumin, slightest possible trace. Sedi- 
ment, 13 hyaline casts, 3 coarse granular casts, all of large diameter; 
a few of the casts with cells adherent; small round cells more numer- 
ous; few squamous cells. 

April 19-20 (sixth day of fast). — Albumin, least possible trace. Sediment, 
15 hyaline casts, 2 coarse granular casts, nearly all of large diameter, 
some with cells adherent; few small round and squamous cells. 

April 20-21 (seventh day of fast) . — Albumin, slightest possible trace. Sedi- 
ment, 5 hyaline casts, 9 coarse granular cells, all of large diameter, 
some with cells adherent; rare normal red blood globule, numerous 
small round cells, few squamous cells; rare spermatozoa, normal in 
appearance. 

April 21-22 (eighth day of fast). — Albumin, slightest possible trace. Sedi- 
ment, 4 hyaline casts, 5 coarse granular casts, nearly all of large 
diameter, some with cells adherent; numerous medium and small 
round cells; rare small caudate cell; little mucus. 

April 22-23 (ninth day of fast). — Albumin, slightest possible trace. Sedi- 
ment, 5 hyaline casts, three of which had a few cells and fat drops 
adherent; 5 coarse granular casts with cells adherent, all casts of 
large diameter; no blood or pus; few small and medium round and 
squamous cells; rare spermatozoa, normal in appearance. 

April 23-24 (tenth day of fast). — Albumin, slightest possible trace (albumin 
cloud more marked than at previous examinations). Sediment, 
8 hyaline casts, some with cells and fat drops adherent; 4 coarse 
granular casts; all casts of large diameter; no blood or pus; few small 
and medium round and squamous cells; little mucus. 

April 24~25 (eleventh day of fast). — Albumin, slightest possible trace (same 
as last examination). Sediment, 4 hyaline casts; 3 coarse granular 
casts; general tendency to diminution in diameter of casts; adherent 
cells and fat drops less numerous; few small and medium round and 
squamous cells. 

300 



MICROSCOPY OF URINE AND TESTS FOR ALBUMIN. 301 

April 25-26 (twelfth day of fast). — Albumin, least possible trace (same as 
last examination) . Sediment, 7 hyaline casts, 6 coarse granular casts, 
chiefly small diameter ; few fat drops and cells adherent ; few leucocytes 
and small round cells; few acid sodium-urate crystals; little mucus. 

April 26-27 (thirteenth day of fast). — Albumin, slightest possible trace 
(reaction less marked). Sediment, 15 hyaline casts; 2 coarse gran- 
ular casts of medium size and with a few cells adherent; few leuco- 
cytes and small round cells. 

April 27-28 (fourteenth day of fast). — Albumin, slightest possible trace 
(same as last examination). Sediment, 6 hyaline casts, 2 coarse 
granular casts, all casts of medium size and with a few cells adherent; 
few leucocytes and small round cells; numerous spermatozoa. 

April 28-29 (fifteenth day of fast). — Albumin, slightest possible trace (very 
faint reaction). Sediment, 15 hyaline casts; 10 coarse granular casts, 
few cells and fat drops adherent; general tendency to decrease in 
diameter of casts; few small and medium round cells; rare small 
caudate and neck-of-bladder cells. 

April 29-30 (sixteenth day of fast). — Albumin, slightest possible trace 
(same as last examination). Sediment, 14 hyaline casts; 12 coarse 
granular casts, chiefly of small diameter; few small and medium 
round cells; rare neck-of-bladder cells. 

April 30-May 1 (seventeenth day of fast). — Albumin, slightest possible 
trace (very faint reaction) . — Sediment, 7 hyaline casts, 2 fine granular 
casts, some with few fat drops and cells adherent; casts of small 
diameter; few small and medium round cells; rare neck-of-bladder 
cells; few squamous cells. 

May 1-2 (eighteenth day of fast). — Albumin, slightest possible trace (very 
faint reaction). Sediment, 8 hyaline casts; 4 coarse granular casts, 
small diameter, few fat drops and cells adherent; few small and 
medium round cells; few squamous cells. 

May 2-3 (nineteenth day of fast). — Albumin, slightest possible trace (very 
faint reaction). Sediment, 8 hyaline casts; 4 coarse granular casts, 
small diameter, few fat drops and cells adherent; few small and 
medium round and squamous cells; little mucus. 

May 3-4 (twentieth day of fast). — Albumin, slightest possible trace (very 
faint reaction). Sediment, 6 hyaline casts; 2 fine granular casts, all 
casts of small diameter, a few containing fat drops; few squamous 
cells; little mucus. 

May 4~5 (twenty-first day of fast). — Albumin, least possible trace (very faint 
reaction). Sediment, 4 hyaline casts; 2 fine granular casts, of small 
diameter; few small and medium round cells; few squamous cells. 

May 5-6 (twenty-second day of fast). — Albumin, slightest possible trace 
(very faint reaction). Sediment, 7 hyaline casts; 4 coarse granular 
casts, all casts of small diameter; few small round and squamous cells. 

May 6-7 (twenty-third day of fast). — Albumin, slightest possible trace 
(very faint reaction). Sediment, 7 hyaline casts; 3 coarse granular 
casts, all casts of small diameter; few small round and squamous cells. 

May 7-8 (twenty-fourth day of fast). — Albumin, least possible trace (very 
faint reaction). Sediment, 6 hyaline casts; 4 coarse granular casts, all 
of small diameter; few small and medium round and squamous cells. 

May 8-9 (twenty-fifth day of fast). — Albumin, slightest possible trace 
(very faint reaction). Sediment, 6 hyaline casts; 2 coarse granular 
casts; all casts of small diameter, occasional fat drops and cells 
adherent; few small and medium round and squamous cells; some 
cells slightly fatty; rare spermatozoa, normal in appearance. 



302 A STUDY OF PROLONGED FASTING. 

May 9-10 (twenty-sixth day of fast). — Albumin, slightest possible trace 
(very faint reaction). Sediment, 8 hyaline casts; 1 coarse granular 
cast; all casts of small diameter; occasional fat drops and cells 
adherent; few small and medium round and squamous cells, some 
cells slightly fatty; rare spermatozoa, normal in appearance. 

May 10-11 (twenty-seventh day of fast). — Albumin, slightest possible trace 
(very faint reaction). Sediment, 8 hyaline casts; 1 coarse granular 
cast; all casts of small diameter, with a few fat drops and cells adher- 
ent; rare spermatozoa, normal in appearance; few medium round 
cells, slightly fatty; few squamous cells. 

May 11-12 (twenty-eighth day of fast). — Albumin, slightest possible trace 
(very faint reaction). Sediment, 6 hyaline casts; 2 coarse granular 
casts; all casts of small diameter, a few fat drops and cells adherent; 
numerous spermatozoa, normal in appearance. 

May 12-13 (twenty-ninth day of fast). — Albumin, slightest possible trace 
(very faint reaction). Sediment, 10 hyaline casts, 3 coarse granular 
casts; all casts of small diameter, few fat drops and numerous cells 
adherent; few small and medium round cells; few squamous cells; 
few spermatozoa, normal in appearance. 

May 18-14 (thirtieth day of fast). — Albumin, slightest possible trace (very 
faint reaction). Sediment, 12 hyaline casts; 1 coarse granular cast; 
all casts of small diameter, few fat drops and rather numerous cells 
adherent; few small and medium round cells; few squamous cells; 
few spermatozoa, normal in appearance. 

May 14~15 (thirty-first day of fast). — Albumin, slightest possible trace 
(distinctly more than last examination). Sediment, 36 hyaline casts; 
2 coarse granular casts; casts of small and large diameter about equal 
in number; few fat drops and epithelial cells adherent; few small 
and medium round cells; numerous squamous cells. 

May 15-16 (first day after breaking fast). — Albumin, slightest possible 
trace. Sediment, 18 hyaline casts; chiefly of small diameter, with a 
few fat drops and rare epithelial cells adherent; few small and medium 
round cells; few squamous cells; very many spermatozoa, normal in 
appearance; many acid sodium-urate crystals. 

May 16-17 (second day after breaking fast). — Albumin, slightest possible 
trace. Sediment, 2 hyaline casts of small diameter; few small and 
medium round cells; few squamous cells; little mucus. 

May 17-18 (third day after breaking fast). — Albumin, slightest possible 
trace. Sediment, 2 hyaline casts; 1 epithelial cast of small diameter; 
few small and medium round cells; few squamous cells; many sperm- 
atozoa, normal in appearance; many calcium-oxalate crystals. 

October 19 (five months after breaking fast). — Albumin, least possible trace. 
Sediment, 2 hyaline casts of small diameter; few small and medium 
round cells; occasional neck-of-bladder cells; few squamous cells. 

SUMMARY. 

The most remarkable change in character of the urine noted during 
the fast was the appearance of albumin and casts on the fifth day, which 
persisted throughout. A summary of the results is given in table 43. 
During the first 24 hours that food was taken, the urine contained sugar 1 
and the sediment showed numerous calcium-oxalate crystals. 

The sexual system of the subject remained active throughout the 
entire period of observation. According to his own statements, a 
seminal emission occurred two nights before the fast began, he had 

^ee p. 292. 



MICROSCOPY OF URINE AND TESTS FOR ALBUMIN. 



303 



Table 43. — Summary of results. 



Day of fast. 


Albumin. 


Hyaline 
casts. 


Granular 
casts. 


Size of casts. 


3d 





1 



13 

15 

5 

4 

5 

8 

4 

7 

15 

6 

15 

14 

7 

8 

8 

6 

4 

7 

7 

6 

6 

8 

8 

6 

10 

12 

36 

18 

2 

2 

2 




3 
2 
9 
5 
5 
4 
3 
6 
2 
2 
10 
12 
2 
4 
4 
2 
2 
4 
3 
4 
2 
1 
1 
2 
3 
1 
2 


1 



Predominating casts large diam. 

Do. 

Do. 

Do. 

Do. 

Do. 
Size diminishing. 

Do. 

Do. 

Do. 
Small diameter. 

Do. 

Do. 

Do. 

Do. 

Do. 

Do. 

Do. 

Do. 

Do. 

Do. 

Do. 

Do. 

Do. 

Do. 

Do. 
Large and small about equal. 
Small diameter. 

Do. 

Do. 


4th 





5th 


SI. possible tr . . . 
do 


6th 


7th 


do 


8th 


do 


9th 


do 


10th 


do 


11th 


do 


12th 


do 


13th 


do 


14th 


do 


15th 


do 


16th 


...do 


17th 


do 


18th 


do 


19th 


do 


20th 


do 


21st 


do 


22d 


do 


23d 


do 


24th 


do 


25th 


do 


26th 


do 


27th 


do 


28th 


do 


29th 


do 


30th 


do 


31st 


. .. .do 




do 




do 




do 




do 







voluptuous dreams without ejaculation on the night of the thirteenth 
fasting day, and a seminal emission on the night of the fifteenth fasting 
day. The urinary sediment contained rare spermatozoa on the third, 
seventh, and ninth fasting days, very numerous spermatozoa on the 
fifteenth fasting day, rare spermatozoa on the twenty-fourth, twenty- 
fifth, and twenty-sixth fasting days, numerous spermatozoa on the 
twenty-seventh, and a few on the twenty-eighth and twenty-ninth 
days. On the first and third days after breaking the fast, during the 
period of extreme mental disturbance, numerous spermatozoa were seen. 
Microscopically the spermatozoa at all times appeared to be normal in 
size and shape. There was no motility, but in every instance the urine 
had been standing for some time before the examination was made. 1 

1 These observations are fully in line with the observations made by Albitsky on fasting rabbits, 
which exhibited sexual excitement after 28 to 30 days of fasting. (Cited by Pashutin, in a Course 
of general and experimental pathology, St. Petersburg, 1902, 2, part 1 .) He also notes that sperma- 
tozoa were frequently found in the urine of starving rabbits. Pashutin likewise cites Manassein, 
who observed spermatozoa motile in the urine of rabbits even 33 hours after the animals had per- 
ished from the want of food, thus indicating the intense persistency of sexual activity. 

In striking contrast to these observations are those of Pojarkov (Compt. rend. Soc. Biol., 1913, 
74, p. 141), who with two dogs which had fasted three months noted profound influence upon 
the sexual activity. Indeed, the depression of the sexual activity was so great and so lasting that 
Pojarkov even suggested that prolonged fasting may result in a bloodless castration. — F. G. B. 



THE RESPIRATORY EXCHANGE. 

Any study of the respiratory exchange in a living animal is of dual 
value, for if properly conducted it throws light upon the character of 
the katabolism and also supplies data for computing by the indirect 
method the heat-production of the body, thus serving as a control 
upon the direct calorimetric measurements of the heat-production. 
In the fasting experiments at Wesleyan University, the measurements of 
the respiratory exchange and of the heat-production were coincidental 
throughout the entire period of the fast. This simultaneous deter- 
mination had certain great advantages, particularly in establishing 
the fundamental laws of metabolism obtaining in the early stages of 
inanition. On the other hand, it gave very little opportunity for 
securing evidence regarding the minimum or basal metabolism of the 
fasting subject, since a prerequisite of a study of the basal metabolism 
is a period of absolute muscular repose. 

In these earlier fasting experiments, such a period of absolute mus- 
cular repose was best secured during the night, when the subject was 
in bed and supposedly sound asleep, i. e., quietly at rest. The subject 
went to bed at 11 o'clock and as the experimental periods were all of 
2 hours' duration, the period from 1 a. m. to 7 a. m. could reasonably 
be taken as the time when the subject had a minimum activity. Unfor- 
tunately no direct evidence regarding the degree of muscular activity 
could be obtained. Furthermore, there was no evidence as to whether 
the subject was asleep the entire time or more or less awake and some- 
what restless. 

It seemed advisable, therefore, in planning the study of the long fast 
at the Nutrition Laboratory, so to arrange the experimental technique 
and routine as to include a series of respiration experiments which 
would throw definite light upon the character of the katabolism, 
measured in both long and short periods, and to be able to isolate 
certain periods in which the subject was perfectly quiet and under the 
same conditions of muscular activity throughout the entire series of 
fasting days. The respiratory exchange of the fasting subject was 
therefore studied in two ways, i. e., by using the bed calorimeter 
and the so-called "universal respiration apparatus." With the bed 
calorimeter both the heat-production and the gaseous exchange could 
be studied throughout the period that the subject remained in the 
chamber. In the experiments at Wesleyan University the subject 
remained in the chamber throughout the whole fasting period and the 
respiratory exchange could be studied in 24-hour periods. In the pro- 
longed fasting experiment at the Nutrition Laboratory, however, the 
calorimeter periods usually began at 9 or 10 p. m. and continued until 
8 a. m. the following day. Thus for 10 or 11 and but rarely for 9 con- 

304 



THE RESPIRATORY EXCHANGE. 305 

secutive hours, the gaseous metabolism of the subject inside the 
chamber could be studied. Graphic records of the degree of muscular 
activity were also secured by a special form of bed. 

The possible differences in the degree of sleep and the degree of 
restlessness and the impossibility of determining the actual periods 
of wakefulness and sleep made it seem undesirable to rely wholly upon 
the determinations of the respiratory exchange made by this method 
for a comparison of the metabolism as the fast progressed. Conse- 
quently each morning, at the conclusion of the night experiment, the 
respiratory exchange alone was studied for two or three experimental 
periods by means of the universal respiration apparatus. With this 
apparatus it was possible to obtain the gaseous metabolism for several 
15-minute periods in which the subject lay perfectly quiet and awake, 
thus giving material for comparison for each day of the fasting period. 
The element of uncertainty as to the degree of muscular activity and the 
degree of wakefulness or sleep was by this method entirely eliminated. 
With both forms of apparatus the respiratory quotient could be accu- 
rately determined, so that a mutual control was obtained on the accuracy 
of the two methods of determining the respiratory exchange. 

APPARATUS AND METHODS USED IN THE CALORIMETER EXPERIMENTS. 

It is unnecessary to enter into the details of the construction and the 
technique of using the respiration calorimeter or of the methods of cal- 
culating the results obtained with it, as these have been fully explained 
elsewhere. 1 Since the publication of this description, however, a number 
of minor though important changes have been made in the apparatus, 
and hence it seems desirable to discuss them here somewhat at length, so 
that the complete technique used in these experiments may be available. 
Although the fundamental principle involved in the determination of the 
respiratory exchange has not been altered in the slightest degree, our 
accumulated experience enabled us to develop a technique to meet the 
special conditions of the fasting experiment which not only facilitated 
the manipulation of the apparatus but also gave greater accuracy. 

ABSORPTION OF WATER- VAPOR AND CARBON DIOXIDE. 

According to the usual arrangement of the absorbing system of 
this apparatus, large porcelain vessels of special form, made by the 
Royal Berlin Porcelain Works of Berlin, Germany, are used to hold the 
sulphuric acid for absorbing the water from the air-current, and silver- 
plated brass cans for holding the soda-lime for absorbing the carbon 
dioxide. Both of these containers weigh considerable and when the 
amount of water or carbon dioxide absorbed is 20 to 30 grams, the 
balance on which they are weighed (which is accurate to about 0.05 
gram) is sufficiently exact. On the other hand, when small amounts of 

1 Benedict and Carpenter, Carnegie Inst. Wash. Pub. 123, 1910. 



306 A STUDY OF PROLONGED FASTING. 

water or carbon dioxide are to be determined, as for example 10 grams 
or less, these vessels are too large. 

In the earlier part of the fasting experiment, the calorimeter experi- 
ments were usually subdivided into three periods, so that the amount of 
carbon dioxide to be weighed represented that produced in about 3 
hours, or approximately 45 to 60 grams. Toward the end of the fasting 
period it seemed desirable to obtain more definite information regarding 
the progress of the metabolism throughout the night and hence an 
attempt was made to secure shorter periods. On one night (May 4-5) 
the periods were but 1 hour long; under these circumstances only about 
11 grams of carbon dioxide were absorbed in each period. It was 
necessary, therefore, to have some form of absorbing vessel which would 
weigh considerably less than the usual containers, as the error in weigh- 
ing might make a measurable difference in the results. We accordingly 
replaced the large vessels with soda-lime bottles and glass sulphuric- 
acid containers, i. e., "Williams bottles," similar to those used in the 
absorbing circuit of the respiration apparatus, and from that time 
divided the calorimeter experiments into relatively short periods 
throughout the night, weighing the water and carbon dioxide in these 
smaller and more accurately weighed containers. A complete descrip- 
tion of the glass soda-lime bottle and the Williams bottle is given else- 
where. 1 Soda-lime was used as the absorbent for carbon dioxide 
throughout the whole fasting experiment, for although at the time the 
description of the respiration calorimeter was published experiments were 
being made with potash-lime, subsequent experience has convinced us 
that, as yet, the original form of soda-lime has not been improved upon. 

ANALYSIS OF CHAMBER AIR AT THE END OF PERIODS. 

While it was desirable to obtain as short and as many periods as 
possible in the night calorimeter experiments, it naturally became diffi- 
cult to arrange the routine so as to secure the largest number of periods 
without decreasing the accuracy and overtaxing the strength of the 
assistants, especially as a continuous metabolism experiment of 35 to 
40 days was quite outside of our experience. With three periods in 
each experiment, as was at first planned, it was possible to arrange the 
program so that a trained observer with two responsible assistants 
could readily carry out the routine of a calorimeter experiment. One 
of the difficult parts of the program was to make provision for the 
analysis of the air residual in the chamber at the end of each period. 
It had been our custom to do this by deflecting a certain volume of the 
air from the outgoing air-current through a series of U-tubes containing 
soda-lime, and pumice stone and sulphuric acid respectively, which 
absorbed the carbon dioxide and water-vapor from the air. The air 

Benedict, Deutsch. Archiv f. klin. Med., 1912, 107, p. 166. See, also, figure 39, p. 316, of 
this publication, for a diagrammatic representation of these bottles. 



THE RESPIRATORY EXCHANGE. 307 

was then passed through a calibrated Bohr gas-meter and returned to 
the system. By this method the amount of carbon dioxide and water- 
vapor in the air residual in the chamber at the end of the period could 
be readily computed. This manipulation of U -tubes, reading of gas- 
meter, barometer, and temperature, all of which was checked by an 
independent observer according to our usual procedure, made an added 
complex at the end of each experimental period, which of itself formed a 
considerable part of the routine. Some simpler method for obtaining 
data regarding the composition of the air inside the chamber was there- 
fore sought. 

Since the publication of the description of the calorimeter, it has been 
our good fortune to become thoroughly familiar with the very ingenious 
and accurate gas-analysis apparatus of Sonden of Stockholm, which 
was used in a research on the composition of outdoor air carried out at 
the Nutrition Laboratory. 1 This apparatus was employed in the fast- 
ing experiment for determining the carbon-dioxide content of the air 
in the calorimeter chamber. By means of a system of previously 
dried gas-sampling tubes, samples of the air could readily be taken in 
a few seconds at the end of each experimental period. The carbon 
dioxide in these samples of air could then be determined on the Sonden 
apparatus the next morning by a skilled assistant. 

Unfortunately, while ideal determinations of the carbon dioxide 
could be obtained by this method, it is no less important to know the 
volume of water- vapor residual in the chamber at the end of the period, 
and the accurate determination of this factor has been one of our most 
perplexing problems. With the U-tube system previously described 
very satisfactory results could be obtained, but with the substitution 
of the gas sampling and the subsequent determination of the carbon 
dioxide on a Sonden apparatus, it was necessary to find some method 
of determining the exact amounts of the water-vapor in the air. It is 
obviously impossible to take a sample of air, even over mercury, and 
retain the water-content for any length of time, as it would either be 
deposited upon the glass tube or so affected that no accurate analysis 
could be made. Furthermore, for analysis by the ordinary gravimetric 
method a sufficient volume of air could not be so stored. We therefore 
began experimenting with a delicate wet- and dry-bulb psychrometer. 

Two mercury thermometers, graduated in 0.01° C, were placed in 
the air-current, one of the thermometers having a moistened strip of 
linen wrapped around the bulb. The depression of the wet bulb was care- 
fully noted and by comparing the observations thus obtained with 
results secured by the U-tube method and particularly with the unique 
and extremely accurate hygrometer of Sonden, 2 it was found that 

Benedict, Carnegie Inst. Wash. Pub. 166, 1912, p. 75. 

2 Sonden, Bihang till K. Svenska Vet.-Akad. Handlingar, 1891, 17, p. 3; see also Meteorolo- 
gische Zeitschr., 1892, p. 81. 



308 A STUDY OF PROLONGED FASTING. 

concordant results could be obtained. Hence, to secure records of 
the water-vapor in the residual air, it was only necessary to place this 
psychrometer in the air-current inside the respiration chamber. It 
was so arranged that the air leaving the chamber came through a pipe 
opening at the rear and extending along the bottom to the front of 
the chamber near the glass window. The air passed over the dry 
bulb of the psychrometer and from there over the wet bulb, and then 
directly to the blower outside. It was therefore possible for the obser- 
ver on the outside to read both mercury thermometers through the 
glass front of the calorimeter chamber; readings to 0.01° C. could 
ordinarily be relied upon. By means of psychrometric tables, the 
amount of water-vapor residual in the chamber could be very readily 
computed. It is thus apparent that by reading the psychrometer and 
taking a single sample of the air in the chamber and subsequently ana- 
lyzing it, it was possible to obtain information regarding the content 
of water-vapor and carbon dioxide in the chamber air at the end of each 
period with a minimum utilization of the assistant's time during the 
night. 

The psychrometer is at present used in this laboratory for short 
experiments with babies and small animals in which respiration 
chambers are employed. Having tested this method of determining 
the water-vapor by two other methods, we felt justified in employing 
it in the long calorimeter experiments, especially as it requires only 
the reading of the two thermometers by the assistant at the end of 
each period. To use the psychrometer successfully, it is necessary 
that the air should pass rapidly over the bulbs of the thermometers. 
Care should also be taken that the fabric about the wet-bulb ther- 
mometer is kept moist, as in long experiments of 10 or 12 hours it 
occasionally becomes dry, so that false readings are obtained. 

It is obvious that no one of these methods, i. e., the sulphuric acid- 
pumice stone, the Sond6n hygrometer, or the psychrometer, gives the 
true value for the water-vapor inside the chamber, as they measure 
only the water- vapor in the outgoing air, and there is certainly an area 
about the ingoing air pipe (where the air is entering absolutely free 
from water-vapor) which is of a much lower water-content, notwith- 
standing the fact that the air is fairly well circulated inside the chamber 
by means of an electric fan. It should be considered, however, that 
the values obtained at the beginning and end of each period are for 
comparison only and we deal here with differences rather than with 
absolute amounts. 

This change in methods was particularly advantageous for the deter- 
mination of the residual carbon dioxide. When U -tubes are used, it 
is necessary to pass 10 liters of air through them in order to secure a 
weighable amount of carbon dioxide. For doing this in a relatively 
short time, such as 3 minutes, a ventilation through the U -tubes of 



THE RESPIRATORY EXCHANGE. 309 

about 3? liters per minute is required and the use of fairly large U-tubes, 
each weighing about 80 to 90 grams. A combination of one soda-lime 
U-tube followed by a pumice-stone sulphuric acid U-tube weighs not 
far from 160 to 180 or even 190 grams, while the amount of carbon 
dioxide to be weighed is sometimes no more than 60 milligrams. When 
experimenting with a man awake or doing severe muscular work, the 
method is perfectly satisfactory, but when experimenting with a man as 
emaciated as our fasting subject, with a minimum metabolism, and 
producing only a small amount of carbon dioxide per hour, it is obvious 
that the residual carbon dioxide in the chamber would be low and it 
would be difficult to obtain very accurate determinations under these 
conditions. With the Sonden gas-analysis apparatus, on the contrary, 
it was possible to determine 0.5 per cent of carbon dioxide to the third 
significant figure with great accuracy. The advantage of thus obtain- 
ing a more delicate determination of the carbon dioxide and at the same 
time decreasing the work required of an assistant during the long night 
period made it desirable to introduce the method by which the sam- 
pling pipette with subsequent analyses could be used to determine the 
carbon dioxide and the wet- and dry-bulb psychrometer for determining 
the water- vapor. The sampling tubes were collected each morning after 
the night experiment was over, and the analyses made on the Sonden 
apparatus by Miss Alice Johnson, whose technical skill in the use of 
this apparatus was well attested in the research on the composition of 
outdoor air previously referred to. 

TENSION EQUALIZER. 

The rubber bathing cap used as a tension equalizer in the earlier 
form of respiration calorimeter has been replaced by a spirometer which 
was first designed for the universal respiration apparatus. A brief 
description of this spirometer with diagram (figure 40) is given on 
page 318, but a more detailed description may be found in an earlier 
publication. 1 

The spirometer is attached directly to the side of the respiration 
chamber and thereby becomes a part of the chamber volume, thus 
providing for fluctuations in the volume of the air inside the apparatus. 
As the carbon dioxide is absorbed by the soda-lime and the oxygen is 
used by the man inside the calorimeter, the total volume of the air 
inside the chamber gradually decreases. Accordingly the spirometer 
bell slowly falls until a certain point is reached where an electric contact 
(not shown in figure 40) is made and oxygen thereby automatically 
admitted by means of an electric valve attached to the oxygen cylinder. 
When sufficient oxygen is admitted to raise the bell and break the 
contact, the flow of oxygen is automatically stopped; in this way the 
supply of oxygen is under continuous control. 

Benedict, Deutsch. Archiv f. klin. Med., 1912, 107, p. 172. See also Benedict and Talbot. 
Carnegie Inst. Wash. Pub. 201, 1914, p. 43. 



310 A STUDY IN PROLONGED FASTING. 

In the fasting experiment, use was made of this spirometer to indi- 
cate the constancy of conditions. For example, if the oxygen supply 
was completely shut off for 5 or 6 minutes before the probable end of 
an experimental period and the pointer on the spirometer bell was 
allowed to rest against the smoked-paper drum, a regular rising curve 
would be drawn on the rotating drum, thus indicating the slow, steady 
fall of the spirometer bell. If the subject made a muscular movement, 
as in turning over, or for any reason there was an irregularity in the 
curve, it was obvious that there was a sudden expansion or contraction 
of the air in the chamber which could not be corrected for by the meas- 
urement of the temperature and barometer. Consequently, if such an 
irregularity in the line occurred during the last 5 minutes of the experi- 
mental period, the length of the period was extended until a regular 
curve could be secured. This was most helpful in many instances. 
Specimen curves are given in figure 36 showing this method of utilizing 
the spirometer. It is of course necessary to note the exact height of the 
spirometer at the moment the period is ended. This is done by read- 
ing the position of the pointer attached to the counterpoise of the spi- 
rometer as it travels over a millimeter scale. 



MAY 1 1. 1912 




8.47 A.M. 



Fig. 36. — Specimen records of change in volume of the spirometer on the bed 
calorimeter during last 5 minutes of periods in experiment with L. 



ARGON IN OXYGEN FROM LIQUID AIR. 

In recent years we have used the nearly pure oxygen obtained from 
liquid air by the Linde Air Products Company. At first we were 
unaware of the fact that the residual gas was not, as commonly con- 
sidered, all nitrogen with an atomic weight of 14, but consisted in large 
part of argon with an atomic weight of 40. Hence it has been necessary 
to emphasize the fact that in computing either the volume of oxygen 
admitted from a cylinder or in calibrating a gas-meter by the method of 
weighing the oxygen, 1 the composition of this residual gas should be 
taken into consideration, as otherwise an appreciable error in the per- 
centage of oxygen may easily occur. Thus, in a series of observations 
carried out on diabetics 2 and likewise another series on muscular work, 3 

Benedict, Physical Review, 1906, 22, p. 294. 

2 Benedict and Joslin, Deutsch. Archiv f. klin. Med., 1913, 111, p 350. 

'Benedict and Cathcart, Carnegie Inst. Wash. Pub. 187, 1913, p. 74. 



THE RESPIRATOEY EXCHANGE. 311 

it was found that this correction for argon in place of nitrogen altered 
the oxygen consumption about 1 per cent and consequently altered the 
values for the respiratory quotient by a like amount. 

GRAPHIC REGISTRATION OF DEGREE OF MUSCULAR REPOSE OF SUBJECT 
INSIDE THE RESPIRATION CALORIMETER. 

The intimate relationship exhibited between the degree of muscular 
repose and the total metabolism compelled us to make sure, at the 
beginning of the fasting experiment, that the measurements of the 
metabolism made with this subject from time to time were comparable 
so far as muscular repose was concerned. Great care was taken to 
secure experimental periods when the subject was perfectly quiet and 
awake, and likewise when he was asleep. All this care would have 
been of no avail, however, if we had not been able to secure periods of 
like muscular repose or activity. If, for example, the subject had been 
noticeably restless on the first few nights of the period of prolonged 
fasting and on the last nights was especially quiet, the decrease in 
the metabolism could not be shown to have been due to the influence of 
the fast, but might have been due to the differences in the degree of 
muscular repose. While statements could be secured from the subject 
or from observers as to how well the subject slept or how quietly he lay, 
no reliance could be placed upon them, as our experience has been that 
such observations are usually untrustworthy. Hence we made use of 
a method of graphic registration, which succeeded the ocular observa- 
tions of the muscular activity used in connection with the experiments 
in Wesleyan University, and likewise had its own development later 
in this laboratory. In the earlier experiments we placed about the 
body of the subject either one or two tube pneumographs in such a 
position that not only was the respiration-rate recorded, but likewise 
any muscular movement of the body. 1 When the subject was lying 
down, it was found that these pneumographs became irksome if worn 
for several hours. In the experiments with diabetics, 2 the use of the 
pneumograph was found to be satisfactory, as the apparatus was 
rarely worn continuously for more than 3 hours. In the fasting experi- 
ment, however, it would be necessary for the subject to wear the pneu- 
mograph for some 12 or 13 hours each night, and as the degree of emacia- 
tion became greater it was quite possible that the discomfort might 
be such as to disturb his sleep if not, indeed, cause pain; also, that the 
subject might turn over during the night and cramp the transmission 
tube in such a way as to prevent proper registration. 

Previous experiences in this laboratory with a suspended cage or 
crib for dogs or infants led us to apply the same principle for devising 
a special form of bed for use in experiments with the universal respira- 

^enedict, Carnegie Inst. Wash. Pub. 77, 1907, p. 10. 

2 Benedict and Joslin, Carnegie Inst. Wash. Pub. 136, 1910, p. 22. 



312 



A STUDY OP PROLONGED FASTING. 



tion apparatus in which the subject lies upon a couch. This bed was 
so suspended that the slightest change in the center of gravity of the 
body, such as moving the hand or the foot, would alter the tension on 
the spring inside a pneumograph and thus transmit the movement to 
a tambour and kymograph. By this means the least muscular activity 
would be recorded. With the suspended crib used in experiments with 
infants, experience has shown that the best point of support was at the 




Fig. 37. — Method for obtaining graphic record of activity in bed calorimeter. 

The subject lies on the bed on the framework inside the calorimeter. One side of the frame 
rests on a knife-edge, K ; the other side is supported by two stout spiral springs, S and S'. Any 
change in the tension on the springs likewise affects the tension on the pneumograph, P, thus 
altering the tension of the air in the pneumograph. By means of a rubber tube and a metal pipe 
passing through the copper wall, C, zinc wall, Z, and asbestos wall, A, of the calorimeter, the lower 
end of the pneumograph communicates with a tambour which writes on the kymograph placed 
above the calorimeter. Any lateral change in the center of gravity of the body instantly produces 
a movement of the pointer on the kymograph. 



THE RESPIRATORY EXCHANGE. 313 

foot of the crib, the spring being placed at the head. With adults, how- 
ever, we soon found that the major movements were lateral rather than 
lengthwise of the body and the supports and springs were accordingly 
placed at the sides of the bed instead of at the head and foot. 

This bed, when used in the calorimeter chamber, was supported at 
one side on two frictionless steel points and at the other by two stout 
spiral springs which could be adjusted by turnbuckles to bring the bed 
to a level position. Obviously any change in the center of gravity of 
the body altered the tension upon the two supporting springs, which 
were therefore elongated or shortened. When the pneumograph was 
attached to the bed, the same force producing the elongation or 
contraction of the springs affected the pneumograph. The change 
in the tension of the air inside the pneumograph was transmitted by the 
usual method, i. e., by means of a metal tube passing through the walls 
of the chamber and subsequently by a rubber tube connecting with the 
tambour, writing-point, and kymograph. The method of obtaining 
this graphic registration of the muscular activity is shown in figure 37. 

In this figure the open end of the bed calorimeter is shown in per- 
spective and in a somewhat schematic way. C, Z, and A represent 
respectively the inner copper wall of the chamber, the zinc middle wall, 
and the outer asbestos wall. The framework of the bed is seen at the 
bottom of the calorimeter chamber with the left-hand edge resting 
on the steel support, K. The two spiral springs, S and S', each pro- 
vided with a turnbuckle, are attached at the upper end to the wall of the 
calorimeter chamber and at the lower end to the right-hand edge of the 
bed. Midway between the springs is attached a pneumograph, P, the 
upper end of which is attached to the wall of the calorimeter chamber. 

The subject, lying upon an air mattress which is in turn resting upon 
a long plate of galvanized iron, is slid on to the bed framework feet first. 
As the weight of the body falls upon the framework, the springs, S and 
S', become extended, the adjustment necessary to secure perfect level- 
ing of the bed being made by means of the turnbuckles. If the subject 
turns during the night, a greater tension is put upon the springs, S 
and S', and the pneumograph, P, is elongated. 1 The air in the tube 
connecting the pneumograph with the outside of the chamber is thus 
somewhat rarefied and the tambour pointer sinks, thus producing a 
depression in the line on the kymograph drum. 

*For the benefit of other workers in this field, it is of interest to record here the recent experi- 
ence of Dr. Paul Roth, of Battle Creek, Michigan. In recording the body-movements of men or 
women lying on beds, he replaced the pneumograph with a small Politzer bulb, so adjusted 
as to be somewhat compressed by the bed-frame. The bulb was connected to the tambour and 
kymograph. Preliminary tests made in the Nutrition Laboratory with the Politzer bulb arrange- 
ment have shown that the results of the variation in pressure on the bulb by variation in muscular 
activity are most satisfactory, not only with adults but also with small animals — a fact of special 
interest in connection with the research on infants. Two serious objections to the pneumograph, 
i. e., the danger of leaks through the rubber and the difficulty of renewing the rubber, are thus 
obviated by the use of this bulb. A flexible rubber bulb of small size is best used. 



314 



A STUDY OF PROLONGED FASTING. 
I. APRIL 10-11. 1912 









t 












M* 














1*0* A.M. 




aS 


























TMS A.M. 










4J* 
















«.!• 


















ill 














M.OC A.M. 




U» ' - 


l.»» * ... 


H*» AJ* 


















ha 


















•!.•» KH. 








I. APRIL 


14-15. 


TlLM P.M. 

1912 






7*9 


, 














1 4*0 AH 


6.IS — 


KB 


















4J0 












rs.00 aj«. 






S3i 
















*—" 


£«4 


LSI 




T&M AM 












iijR 




— ' 














ILM AJ4. 














11.13 


IO20 



-TM7AK HI. APRIL 29-30. 1912 



SA7 








Ul 


t&OOAM. 


~~ * 




xso t — ■ 


f 


U>« " ' 










1.10 




11.17 AM. ' 


IU» 


IOJ5 






r — 



EZ. MAY 13-14. 1912 



T7.3I AM. 



J~ 



127 
rJ*. 



•XT 



tZ/JI AJ*. 



fx. 



ILS2 

•uUr 



+- *" 



Fig. 38. — Specimen pneumograph records of movements of bed calorimeter lever mattress 
support in night experiments with L. 



THE RESPIRATORY EXCHANGE. 315 

The apparatus is extremely sensitive and shows plainly such minor 
muscular motions as movements of the hand or arm to one side or, 
indeed, the twisting of the feet. Throughout his whole stay in the 
laboratory, L. slept on this bed inside the respiration chamber each 
night and a kymograph record was therefore obtained for the whole 
period. The drum of the kymograph was usually rotated at such a speed 
as to give one revolution of 500 mm. per hour. 

Although these kymograph records were not secured primarily for 
publication, four typical records have been arbitrarily selected for 
reproduction in figure 38. The curve for April 10-11, 1912 (Curve I), 
which represents the record obtained on the first night which the sub- 
ject spent inside the chamber, begins at 11 p. m., April 10, and ends at 
8 h 02 m a. m., April 11. For the most part the subject was remarkably 
quiet, showing no considerable degree of restlessness until about 5 a. m. 
on April 11. 

The curve for April 14-15, 1912 (Curve II) represents the record 
obtained on the first night of the fasting period. The line here is 
remarkably regular, showing relatively few major muscular movements. 
When there is a distinct change in the level of the mark made by the 
pointer, as is seen at approximately 5 h 12 m a. m., this is an indication 
that the subject changed his position, probably turning on his side. 

Another curve (Curve III) has been selected from those obtained 
about the middle of the fasting period, which represents the record for 
April 29-30, 1912. This also shows a general regularity of line, with 
occasional indications of changes in position. 

The last curve (Curve IV) was obtained on the next to the last night 
of the fasting period, May 13-14, 1912. Between 12 h 45 m a. m. and 
l h 45 m a. m. there was considerable movement for a few moments at 
two or three different times, but otherwise the record has much the 
same characteristics as the other curves shown. 

From these curves and also from other curves which it is imprac- 
ticable to reproduce here, we may logically infer that this subject 
was particularly quiet inside the respiration chamber. While he did 
not sleep the entire time, yet the kymograph records show that he 
was for the most part very comfortable inside the chamber — indeed, 
he repeatedly made the statement that he was very comfortable 
throughout the night. It is of interest to note that at the hospital 
he said that the bed there was not so comfortable as the bed inside the 
calorimeter chamber. 1 

METHODS USED IN EXPERIMENTS WITH THE RESPIRATION APPARATUS. 

The universal respiration apparatus, which was used for the respira- 
tion experiments throughout the day, is based upon the same principle 
as the large respiration calorimeters. A great variety of experiments 

x See page 51. 



316 



A STUDY OF PROLONGED FASTING. 



have been made with it in this laboratory, and its accuracy has been 
thoroughly tested. Not only men and women have been used as sub- 
jects, but, by adding a small chamber, experiments have also been 
made with infants and small animals. The apparatus has been 
described in detail elsewhere. 1 

With this apparatus the subject lay quietly on the same bed upon 
which he slept during the night, the bed being withdrawn from the 
respiration chamber and placed upon a small framework in the calo- 
rimeter laboratory. He was covered with bed clothing and two soft- 
rubber nose-pieces were inserted in the nostrils, the subject being cau- 
tioned to keep his mouth closed. After he had breathed a few minutes 
through a two-way valve opening into the room, the valve was turned 
and he began to breathe into a closed volume of air — some 8 or 10 
liters — which was kept in motion by a ventilator or blower. As the 
air left the nostrils of the man, it was carried by the blower to suitable 




Fig. 39. — Schematic outline of universal respiration apparatus. 

The subject, lying upon a couch or bed, breathes either through the two nose-pieces or a mouth- 
piece into a ventilating current of air, kept in motion by a rotary pump. The moisture in the 
air is absorbed in two glass vessels containing sulphuric acid, an empty glass vessel serving as a 
trap to prevent accidental back suction of acid. The dried air then passes through soda-lime and 
again through sulphuric acid in a special form of bottle and finally through a can containing 
sodium bicarbonate to free the air of any traces of acid vapor. Oxygen is introduced as desired. 
The air is then ready to be inhaled by the lungs. As the air leaves the lungs, the changes in the 
volume of the confined air are recorded on the spirometer, which moves freely up and down with 
each inspiration and expiration. The change in weight of the soda-lime vessel and its accompany- 
ing sulphuric-acid bottle gives the weight of the carbon dioxide produced. The weight of the 
oxygen is obtained either by noting the loss in weight of the cylinder of the gas or measuring the 
gas carefully admitted through a meter. 

Benedict, Deutsch. Archiv f. klin. Med., 1912, 107, p. 156. 



THE RESPIRATORY EXCHANGE. 317 

containers in which the water and carbon dioxide were absorbed; 
oxygen was next added from a cylinder of weighed gas or through a 
calibrated meter to replace that used by the man; the air was then 
returned to the subject. The amount of carbon dioxide excreted was 
obtained from the changes in weight of the absorbers and the amount 
of oxygen consumed from the record of the oxygen admitted to the air- 
current. Experiments could be made with this apparatus with periods 
as short as 15 minutes. The general scheme of the respiration appa- 
ratus is shown in figure 39. 

In this apparatus provision has been made for fluctuations in the 
volume by attaching a tension equalizer. In the earlier forms of the 
respiration apparatus, a rubber bathing cap was used as a tension 
equalizer, but more recently this has been replaced by a spirometer. 
This spirometer not only provides for the fluctuations in the volume of 
air, but has been utilized for recording the character of the respiration, 
as has already been noted in a previous section of this publication. 
(See page 158.) It has also been used with the bed calorimeter for indi- 
cating the constancy of conditions. (See page 310.) The details of 
the spirometer are shown in figure 40. 

With each inspiration and expiration, the thin copper bell, c, of the 
spirometer falls and rises in the annular space between the copper walls, 
a and b, this space being filled with water. To the counterweight rod, 
g, g, g, is attached a pointer, h, which writes on the smoked-paper 
surface of the kymograph drum. A wheel, r, with a milled edge, is 
rotated by each upward movement of the cord, t, which rests in a groove 
in the edge of the wheel, the pawl, u, preventing any backward move- 
ment of the wheel. By means of a platinum contact on the periphery 
of r, each complete revolution of the wheel may be recorded. An expla- 
nation of the use made of the records obtained with this spirometer 
and a series of typical kymograph curves are found on pages 158 to 160. 

The respiration apparatus was used regularly each morning of the 
fast for an experiment immediately following the night's sojourn in the 
bed calorimeter. The subject, lying upon his bed, was transferred 
directly from the calorimeter chamber to the respiration apparatus. 
He then turned upon his side and urinated without rising and the 
respiration experiment was begun shortly afterwards. In the latter 
part of the fast, the apparatus was used for a respiration experiment 
each evening about an hour before the subject entered the calorimeter. 
At irregular intervals throughout the fast the respiratory exchange was 
also studied with the respiration apparatus, while the subject was 
sitting quietly in a chair or writing steadily, as he did much of the time. 
The apparatus was likewise used for experiments in which the subject 
breathed an oxygen-rich atmosphere while lying upon the couch. 

All of these experiments included two or three periods of approxi- 
mately 15 minutes each, in which records of the degree of muscular 



318 



A STUDY OF PROLONGED FASTING. 




Air leaving the lungs enters the pipe, m n, 
and passes into the spirometer bell, c. This 
bell moves up and down in the annular space 
filled with water between the inner copper 
shell, a, and the outer copper shell, b. The 
bell is counterpoised by a rod, g g g, and a 
small supplementary weight, I, attached to a 
cord, t, the counterpoise being «upported by 
an aluminium wheel, e, over which the thread 
d passes. The height of the spirometer can 
be read by the pointer, h, on a millimeter 
scale, p, or the pointer h can be made to 
write directly upon a kymograph drum. A 
work adder wheel, r, is so attached that the 
cord t passes through a groove in its peri- 
phery, and the wheel conse- 
quently rotates with each down- 
ward motion of the bell c. The 
pawl, m, prevents back motion. 
A projection, w, in the periphery 
of the wheel makes an electric 
contact and permits graphic re- 
cord of each complete revolu- 
tion of the work adder wheel. 

For counting the respirations 
a new attachment, s, has been 
employed. A steel wire is loosely 
fastened around the hub of the 
wheel, r. By means of a light 
hair spring, lateral tension is 
brought against 
the movable rod, g. 
As the rod ascends, 
the steel wire ro- 
tates out of the 
mercury cup, 
breaking the elec- 
tric contact; as the 
rod moves down, 
the contact is 
made. A small 
stop above the 
mercury cup pre- 
vents the wire from 
rising too far. 



Fig. 40. —Spirometer for studying thejnechanics of ventilation. 



THE RESPIRATORY EXCHANGE. 319 

repose were obtained by means of the special form of suspended bed, 
and the pulse-rate was regularly observed. 1 The subject L. adjusted 
himself very readily to this apparatus, finding it not at all uncomfort- 
able. Indeed, on one or two occasions he expressed his enjoyment of 
the soothing sensation produced by the slight sound of the blower. 
He seemed to be in no wise affected by the apparatus and to him appar- 
ently the respiration experiment was but a slight incident in the day's 
program. These experiments were all personally supervised by Mr. 
Carpenter, who found this subject more nearly ideal than any subject 
he had ever studied, as the man could be relied upon to keep absolutely 
quiet throughout the whole period. The record of the degree of mus- 
cular repose also shows this fact, 2 and it is especially advantageous to 
have this assurance that, as the fast progressed, whatever disturbance 
in the total metabolism is observed as a result of varying body position 
or the inhalation of oxygen-rich atmospheres, it certainly was not 
complicated by extraneous muscular activity. 

Aside from the value of being able to study the respiratory exchange 
under the different conditions of waking, sitting, and writing, the 
universal respiration apparatus offered special advantages for likewise 
studying the mechanics of respiration, including the respiration-rate, 
the character and volume of each respiration, and the ventilation of 
the lungs per minute, and for indicating in general any abnormality in 
the mechanics of respiration. It was also possible to determine the 
alveolar air of the subject in these respiration experiments. 3 

As both the morning and evening series of respiration experiments 
were made under the same conditions of muscular repose and with the 
subject awake, the results obtained are perfectly comparable, so that 
they give excellent data for drawing sharp conclusions as to the influ- 
ence of prolonged fasting upon the general metabolism. Furthermore, 
since two wholly independent series of respiration experiments were 
obtained with different apparatus and at a different time of day, the 
individual periods of both the experiments with the bed calorimeter 
and the respiration apparatus are made doubly valuable by this check. 
It is important to bear in mind, however, that the experiments with the 
universal respiration apparatus gave no evidence regarding either the 
water-vapor exhaled from the lungs and skin or the cutaneous respiration. 
As Magnus-Levy has pointed out, 4 there is a greater amount of carbon 
dioxide excreted through the skin than of oxygen absorbed, so that 
there is a tendency for the respiratory quotient to be affected by about 
0.01. When, therefore, a respiratory quotient of 0.73 is obtained with 
a subject in the respiration chamber of the calorimeter, a respiratory 

^ee pp. 311, 99, and 110. 

2 See figure 22 on page 159. 

3 See discussion of these results on pages 168 to 181. 

4 Magnus-Levy, von Noorden's Handbuch der Pathologie des Stoffwechsels, 1896, 1, p. 218. 



320 A STUDY OF PROLONGED FASTING. 

quotient of but 0.72 would, under like conditions, be obtained with the 
respiration apparatus. On the other hand, the experiments made 
with the universal respiration apparatus are extremely helpful as a 
general index of the respiratory exchange from which the calorimetry 
can be computed by the indirect method. We considered it of im- 
portance to make a special effort to secure experiments of short 
duration, as the technique of the experiments made by Luciani on Succi 
have been adversely criticized by Zuntz and the experiments made by 
Zuntz and his co-workers on the fasters Breithaupt and Cetti in Berlin 
were certainly complicated by colic and a head cold. Moreover, it is 
not unreasonable to suppose that the technique in thirty years has 
been materially improved. 

STUDIES WITH THE BED CALORIMETER. 
ATMOSPHERIC CONDITIONS INSIDE THE CHAMBER. 

Prior to a consideration of the results of the study of the gaseous 
exchange inside the bed calorimeter, it is advisable to note the exact 
conditions of ventilation, temperature, and particularly relative 
humidity under which this subject was living in the chamber. The 
respiration calorimeter was ventilated at a rate of approximately 40 
liters per minute, or roughly speaking, 2,400 liters per hour. Since 
the volume of the chamber was not far from 800 liters, theoretically 
the air would be replaced inside the chamber three times each hour. 
The cooling arrangement prevented an abnormal rise in the tempera- 
ture, and a study of the relative humidity under these conditions pre- 
sents certain features of interest. Since the air is dried over sulphuric 
acid before it is returned to the calorimeter, it enters the respiration 
chamber absolutely water-free and consequently the only sources 
of water-vapor inside the chamber are the lungs and the skin of the 
man. The ventilation of the chamber per hour, the amount of water 
vaporized per hour, the average temperature of the calorimeter cham- 
ber, and the relative humidity of the air are given in table 44. 

Daily tests, in which the number of revolutions of the blower are 
recorded by an automatic counter, have shown that in general 210 revo- 
lutions of the blower correspond to a ventilation of 1 cubic foot or 28.315 
liters of air. The total ventilation may therefore be obtained by divid- 
ing the number of revolutions by this factor and multiplying by 28.315. 
As a matter of fact, the number of revolutions per cubic foot of air 
was determined each day and this variable used in the calculation. 
From the length of period as given in table 44, the ventilation per 
hour was readily found. 

All of the water-vapor removed from the chamber was absorbed by 
sulphuric acid as the ventilating current passed through the absorbing 
system. The total amount was corrected for the small amount of 



THE RESPIRATORY EXCHANGE. 



321 



water vaporized from the wet bulb of the psychrometer, and the amount 
per hour found in the usual way. 

The average calorimeter temperature was secured by means of a 
series of resistance thermometers. The relative humidity was calcu- 
lated from the amount of water vaporized per liter of ventilation and 
the number of milligrams of water-vapor in one liter of air saturated 
at the calorimeter temperature. 

The rate of ventilation and the rate of carbon-dioxide production 
were such that the usual proportion of carbon dioxide residual in the 

Table 44. — Ventilation of chamber and relative humidity during experiments with L. in bed 

calorimeter at night. 



Date. 



Day of 

fast. 



Period. 



Ventila- 
tion per 
hour. 



Water 
vapor- 
ized per 
hour. 

B 



Average 
temper- 
ature 
of the 
chamber. 
C 



Relative 
humid- 
ity. 



1912. 

Apr. 10-11 

11-12 

12-13 

13-14 

14-15 

15-16. . . . 
16-17. . . . 

17-18 

18-19 

19-20 

20-21 

21-22 

22-23 

23-24 

24-25 

25-26 

26-27. . . . 

27-28 

28-29 

29-30 

Apr. 30-May 1 

May 1- 2 

2-3.... 
3-4.... 
4-5 



5- 6. 

6- 7. 

7- 8. 

8- 9. 
9-10. 

10-11. 
11-12. 
12-13. 
13-14. 
14-15. 
16-17. 
17-18. 



1st. 

2d.. 

3d.. 

4th. 

5th. 

6th. 

7th. 

8th. 

9th. 
10th. 
11th. 
12th. 
13th. 
14th. 
15th. 
16th. 
17th. 
18th. 
19th. 
20th. 
21st. 
22d.. 
23d.. 
24th. 
25th. 
26th. 
27th. 
28th. 
29th. 
30th. 
31st. 



ll h 
10 
11 
11 

9 

9 

9 

8 

9 



38 m p.m. 

13 p.m. 

13 p.m. 

09 p.m. 

30 p.m. 

41 p.m. 

22 p.m. 
58 p.m. 
57 p.m. 

30 p.m. 

23 p.m 
47 p.m, 

50 p.m, 

12 p.m, 
18 p.m, 

40 p.m, 

53 p.m, 

13 p.m, 
27 p.m. 

46 p.m. 

29 p.m. 
15 p.m, 

51 p.m. 
35 p.m, 

31 p.m, 

14 p.m. 
35 p.m. 
34 p.m, 

41 p.m. 
08 p.m, 

47 p.m. 
55 p.m. 

30 p.m. 
01 p.m. 
38 p.m. 

54 p.m. 
57 p.m. 



to8 h 



02 m a.m. 

00 a.m. 

00 a.m. 

00 a.m. 

00 a.m. 

00 a.m. 

55 a.m. 

00 a.m. 

50 a.m. 
00 a.m. 

58 a.m. 
55 a.m. 
00 a.m. 
52 a.m. 
00 a.m. 
52 a.m. 
00 a.m. 
00 a.m. 

51 a.m. 
47 a.m. 
00 a.m. 
50 a.m. 

00 a.m. 

55 a.m. 

01 a.m. 
50 a.m. 
46 a.m. 
50 a.m. 

59 a.m. 
03 a.m. 
00 a.m. 
30 a.m. 

30 a.m. 

31 a.m. 

56 a.m. 
54 a.m. 
54 a.m. 



liters. 
2253 
2472 
2473 
2454 
2474 
2468 
2469 
2369 
2328 
2410 
2467 
2432 
2395 
2204 
2347 
2254 
2376 
2313 
2134 
2321 
2309 
2334 
2307 
2356 
2268 
2212 
2137 
2195 
2205 
2309 
2243 
2312 
2318 
2257 
2395 
2318 
2455 



gm. 
25.3 
26.3 
26.6 
27.1 
22.8 
25.6 
28.7 
22.8 
21.1 
19.4 
18.9 
19.2 
21.1 
17.0 
18.3 
17.9 
18.4 
17.6 
13.6 
15.9 
15.6 
16.0 
15.5 
15.8 
14.6 
15.7 
16.7 
16.5 
17.5 
18.0 
18.7 
18.9 
19.3 
19.7 
17.9 
20.0 
23.7 



°C. 
21.32 
20.17 
21.28 
20.72 
20.77 
20.65 
21.41 
20.77 
21.19 
20.09 
20.76 
20.37 
21.28 
20.82 
20.91 
20.75 
21.16 
20.58 
20.08 
20.08 
20.19 
20.65 
20.23 
20.57 
20.73 
20.68 
20.77 
20.54 
20.64 
20.53 
20.71 
21.02 
21.25 
21.12 
21.10 
20.65 
20.46 



p. ct. 
62 
62 
59 
63 
52 
59 
63 
54 
50 
47 
43 
46 
48 
44 
44 
45 
43 
43 
38 
40 
39 
39 
39 
38 
38 
40 
45 
44 
46 
46 
48 
46 
46 
49 
42 
50 
56 



322 A STUDY OF PROLONGED FASTING. 

chamber at the end of each experimental period throughout the night 
was not far from 0.4 per cent by volume or approximately 13 times that 
of normal air. This percentage of carbon dioxide in the air, and indeed 
a very much higher percentage, has been shown to be entirely without 
effect upon persons breathing such an atmosphere, 1 so that it may be 
stated with perfect confidence that the excess amount of carbon dioxide 
present in the chamber could in no way have influenced either the 
respiratory exchange or the heat-production of the subject. 

As will be seen from table 44, the ventilation of the chamber aver- 
aged not far from 2,200 to 2,300 liters per hour throughout the 31 days 
of the experiment, the range being from 2,134 liters to 2,474 liters per 
hour. The hourly vaporization of water had a tendency to decrease as 
the fast progressed, the largest amount being on the night of the third 
day of fasting and the smallest on the night of the fifteenth day of 
fasting. The average temperature of the calorimeter remained for the 
most part within a few tenths of a degree of the average figure, 20.6° C. 

The relative humidity shows an interesting course. Beginning with 
approximately 60 per cent on the nights following food, it decreased 
to a minimum level of approximately 39 per cent from the fifteenth to 
the twenty-first day and thereafter rose gradually to the end of the 
fast. The variations in the excretion of water-vapor and the cause of 
the fluctuations in the relative humidity will be discussed in a subse- 
quent section of this publication. 2 The results secured in the measure- 
ments of the respiratory exchange inside the bed calorimeter may, 
therefore, now be considered. 

MEASUREMENT OF THE RESPIRATORY EXCHANGE INSIDE THE 
BED CALORIMETER. 

With the bed calorimeter it is possible to determine simultaneously 
the water vaporized inside the chamber, the carbon dioxide produced, 
and the oxygen consumed. These determinations were made directly 
on four nights prior to the fast, on the 31 nights of the fast, and for two 
nights after the fast. While the greatest emphasis must be laid upon 
the total amounts measured, the absorbing vessels were changed several 
times during the night, so that the experiment was usually subdivided 
into three periods. In the latter part of the fast, the measurements 
were made in five or six periods, and on two nights, seven and nine 
periods respectively. It was accordingly possible to compute the car- 
bon-dioxide output, the oxygen intake, and the respiratory quotient for 
the whole experiment and also for the individual periods, thus giving a 
control on the measurement of the respiratory exchange. 

Benedict and Milner, U. S. Dept. Agr., Office Exp. Stas. Bui. No. 175, 1907, p. 237. 
2 See page 373. 



THE RESPIRATORY EXCHANGE. 



323 



Periodic Changes in the Metabolism. 

The average results for the experimental periods have been plotted 
in the form of curves for each night that the subject spent inside the 
respiration chamber. (See figures 41 to 44.) These results are given 
in cubic centimeters per minute, the scale values on the outside indicat- 
ing the values for the oxygen consumed, and those on the inside the 
values for the carbon dioxide produced. The respiratory quotient for 
each individual period is placed between the oxygen and the carbon- 
dioxide curves. Thus, on the night of April 10-11, the oxygen intake 
for the first period averaged 284 c.c. per minute, for the second period 



! :00P.M. 10.00 I2;00 2:00A.M. 4:00 6:00 8:00 



230 
.220 



tre 

26C 

2JC. 

24C .230 
.220 
.210 

280 . 

270 . 

260 . 

2S0 . 

240 . 

230 230 
220 
210 
200 



190 
180 



0.84. 0.81 



APR. 13-14 



CO, 



00 P.M. 10.00 12-.0Q 2:00A.M. 4:00 6:00 8:00 



Oj 
830 


co a 

-180 . 

o« 


APR. I4--I5 









220 


0.73 


210 




0.78 






190 


_170 

.160 

C0 2 








0.77 












220 


°i 


APR. 15-16 




<7\ 


210 


0.78 




.170 

.160 

C0 2 


0.74- 








.ISO 


210 


°i 




APR. 16-17 




(5) 


200 


.155 

eog 






0.75 




190 


0.71 










.145 




210 


• °! 




APR. 17-18 


. 


200 
190 


C0 2 


0.73 




0.7 S 


180 




0.76 








.140 













Fig. 41. — Curves showing oxygen consumption, 
carbon-dioxide production, and respiratory 
quotient during night periods in the bed 
calorimeter for the four days preceding the 
fast and the first to the fourth days of the 
fast. 



324 



A STUDY OF PROLONGED FASTING. 



265 c.c. per minute, and for the third period 270 c.c. per minute. The 
carbon-dioxide production for the corresponding periods was respec- 
tively 227, 221, and 218 c.c. per minute and the respiratory quotients 
0.80, 0.84, and 0.81 respectively. 

The subdivision into experimental periods was made in an attempt 
to secure information regarding the periodic changes throughout the 
night. But from fundamental factors in the technique of the calori- 
meter experiments, the longer the experimental periods, the more accu- 
rate are the measurements of the carbon-dioxide production and es- 
pecially of the oxygen consumption; hence by subdividing these total 
values, a certain degree of accuracy is sacrificed in the measurements, 
although the average values for the night are unaffected. This may 
explain certain discrepancies in the respiratory quotient and in the 
general conformity of the curves for the oxygen and the carbon dioxide. 
5 It will be seen that as a rule the curves for the carbon dioxide and 
oxygen are essentially parallel, although they are by no means straight 

8:00 P.M. ifcQO IfcQO 2O0A.M 4 00 8:00 6:00 BjOOP.M I0:0Q 12:00 2:OOA.M *:O0 6.00 8:00 



°, 


o« 











180 
170 

180 
170 
160 

180 
170 

I7S 
165 

175 
165 
155 

170 
16C 


CO, APR. 23-24. 

140 °l 1 


0.75 


% 




0.°* co « 


0.7 Z 


APR. 18-19 


0.7 e 

130 CO, 


071 










190 . 




0.78 














°» 






G 












vii 


170 


0.77 






0.70 

.130 CO, 




APR. 24-25 




I3S 


0.74 


















0.74 




200 


o, 








© 

■20 








0.70 


APR. 19- 


-1ZO 






190 


_140 

co » 


0.67 






ISO 








« APR. 25-26 








<S 








0.74 








CO, 07 * 






















H 






0, 


APR. 20- 




G\ 


-125 






200 










21 


0.70 






190 


°z 




APR. 26-27 




<3 




-140 

CO, 


0.71 


L.30 CO, •»* 






ieo 




0.73 




0.71 






.120 














.130 
















°» 






<5 




°* 






© 








0.69 
.130 

.120 C0 » 




APR. 27-28 






.140 CO, 


0.70 




190 


APR. 


21-22 


0.75 




180 








0.73 

1 






0.73 

1 


0.77 








.110 








.130 




J 


















°2 




APR. 22-23 


© 


APR. 

°2 


28-29 


' 0.71 


C 




.145 

- 135 CO, 


0.72 


1 


1 










I7C 


0.81 




.125 

CO, °- 72 
















1 


r 






113 




























— 



F IO- 42. Curves showing oxygen consumption, carbon-dioxide production, and respiratory quotient durir 

night periods in the bed calorimeter for the fifth to the fifteenth days of the fast. 



THE RESPIRATORY EXCHANGE. 



325 



lines for the whole experiment. Occasionally, discrepancies are found, 
as on April 10-11 (the first night of the experiment) when the carbon- 
dioxide production is higher in the second period than in the last period, 
while as a matter of fact the oxygen consumption is apparently some- 
what lower. This is especially noticeable on April 22-23, when the 
carbon-dioxide production in the last period increased and the oxygen 
consumption decreased. As would be expected, the possibilities for a 
discrepancy between the curves increase as the period is shortened and 
consequently we find on the night of May 4-5, when the experiment was 
divided into nine periods, that while the values as a whole are approxi- 
mately parallel, in the sixth and seventh periods there is a great in- 
crease in the oxygen consumed which is unaccompanied by a corre- 
sponding increase in the carbon-dioxide production. Similar irregu- 
larities are to be noted on the night of May 13-14. In general, however, 
there is a striking tendency toward parallelism in the two curves. 

After the first three nights of fasting, the minimum values for carbon 
dioxide and oxygen are usually found in the middle period of the night, 
i. e., from 2 to 4 a. m., or thereabouts, the morning period almost inva- 



J:00P.M. 


10:00 


12:00. 2:00A.M. 4:00 


6:00 


8:00 


1 t 
. 

CO, 
.0 .125 

.115 


0* 




APR. 29-30 




Q 


C0 2 


0.69 


0.72 


0.72 












1 i0. 
1 .120 
.110 


°l 




*PR.30-MAY 1 


to 


co 2 


0.70 


0.73 


0.73 












1 5 . 
Ii5 .125 
.115 


o, 




MAY 1-2 


1 0.75 


O 


co 2 


0.70 


0.7Z 

1 


| 










1 ». 
1 <0 .115 
.I0S 


o. 




MAY 2-3 


L_ 


(3) 


COj 


0.70 


0.70 


0.7* 














! ». 
>0 .115 
-!<35 


o, 


MAY 3-4. 






© 


CO, 


0.70 


0.71 


0.73 




r~ 











6.00 P.M. 10.00 12. 00 2:00A.M. 4:00 6:00 



I :o. 43. — Curves showing oxygen consumption, carbon- 
dioxide production, and respiratory quotient during , 50 
night periods in the bed calorimeter for the sixteenth 
to the twenty-fourth days of the fast. 




326 



A STUDY OF PROLONGED FASTING. 



riably showing a tendency to rise. This may be seen with great regu- 
larity throughout most of the curves, although there are enough 
exceptions (for instance, on May 4-5 and May 6-7) to make it inappli- 
cable in all cases. 

After the first few days of the fast, one would not expect a great 
change in the respiratory quotient, since there would be no material 
alteration in the character of the material oxidized in the body. An 
examination of the respiratory quotients given with the curves shows 
that they run not far from a constant value throughout the night with 
the different conditions of food and fasting. Thus, on the first few 
nights with food, the values are considerably above 0.80, but with the 
beginning of the fast they drop rapidly to about 0.74, remaining not 
far from 0.72 throughout the remainder of the fast. Occasionally cer- 
tain fluctuations above or below the average figure may be observed, 



8:00P.M. 10:00 12:00 f.QQAM. 4:00 6:00 



CO, 



115 

105 



125 
115 
105 



iL. 



MAY 9-10 



© 



may io-ii 



MAY 11-12 



CO, Q.75 



MAY 12-13 



© 



>0 P.M. 10:00 12:00 2:00 A.M. 4:00 6jOJJ 9:00 



200 
190 
180 



CO, 



120 
110 



200 
190 
180 



135 co z 

125 

IIS 



"MAY 16-17 



0.80 0.84 





MAY 13-14 






—1 . 




itt 1 r^ 




C0 2 


— 1 ,. 0.74 n«7 




















L 










o, 






MAY 14-15 




049 






~L 


0.73 








C0 2 


0.70 






0.72 074 











© 



Fig. 44. — Curves showing oxygen consumption, carbon-dioxide production, and respiratory quotient during nigfc 
periods in the bed calorimeter for the twenty-6fth to the thirty-first days of the fast and the secon 
and third food days. 



THE RESPIRATORY EXCHANGE. 327 

but these may easily be attributed to the fact that the shortness of the 
period affected the determinations of the oxygen consumption. It may 
be considered an established fact that respiratory quotients more than 
0.02 above or below the average value for the night are due to acci- 
dental variations in the determinations. A low respiratory quotient 
following a high quotient may frequently be noted, showing that there 
is a compensation in the measurement of the oxygen consumption as 
the experiment continues. 

Inasmuch as the determination of the respiratory quotient requires 
extremely accurate determinations of both the oxygen consumption 
and the carbon-dioxide production, it is obviously very much more 
difficult to secure accurate respiratory quotients than accurate measure- 
ments of either the carbon dioxide or the oxygen. Accordingly, while 
we feel that it is legitimate to accept the values for the carbon dioxide 
and even for the oxygen for short periods, we are by no means certain 
that we are justified in laying considerable stress upon the respiratory 
quotients in periods so short as those in the calorimeter experiments. 

A careful scrutiny of all of the kymograph records shows that the 
extraneous muscular movements, although not absolutely constant in 
every period, are so slight that they may be practically neglected. 
Such movement as there was did not correspond closely to the general 
trend of the katabolism, for although the subject was more active 
during the morning period, the activity was not sufficient to account 
for the great difference in the katabolism. On the other hand, it is per- 
fectly clear from his own records that the subject was usually in deep 
sleep in the middle of the night, as he often reported in the morning 
that he awoke about 4 a. m. and lay awake until the end of the experi- 
ment. A relationship between deep sleep and the metabolism is there- 
fore indicated, a relationship which will be discussed in a subsequent 
section. 

Perhaps the most striking fact shown by the whole series of curves 
is that the subject had by no means a constant metabolism. This man 
was living on a low metabolic plane, had a remarkable degree of mus- 
cular repose as shown by the kymograph records, and was without 
food in the alimentary tract ; and yet, as has already been pointed out, 
the curves show a distinct tendency to fall off in the first part of the 
night until a minimum is reached from 2 to 4 a. m., and then to rise 
again in the later morning hours. 

Total Metabolism. 

The difficulties incidental to comparing the short-period values for 
the gaseous exchange are eliminated when one uses as a unit the 
results obtained during the entire sojourn of the subject inside the 
respiration chamber during any given night — that is, for a period of 



328 



A STUDY OF PROLONGED FASTING. 



10 or 12 hours. Consequently a comparison may be made of the 
results obtained for the individual nights as the fast progressed. We 
may, indeed, go further and compare not only the average values found 
throughout the night, but likewise the average values for the minimum 
periods in the experiments. While obviously there is an opportunity 
for possible error in thus selecting minimum periods, particularly in 
the measurements of the oxygen consumption, nevertheless it is be- 
lieved that such errors will be equalized throughout a 31 -day fast. 
Accordingly, in table 45 we give both the average and the minimum 

Table 45. — Gaseous exchange of subject L. during experiments in the bed calorimeter at night. 







Carbon dioxide 


Oxygen per 








Date. 


Day of 
fast. 


per minute. 1 


minute. 1 


Respira- 
tory 
quotient. 

(A-C) 


Average 
pulse- 
rate. 


Average 

body- 
tempera- 
ture. 


Aver- 
age. 


Mini- 
mum. 2 


Aver- 
age. 


Mini- 
mum. 2 






A 


B 


C 


D 


E 


F 


G 


1912. 




ex. 


c.c. 


c.c. 


c.c. 






°C. 


Apr. 10-11 




224 
228 
218 
180 
165 


218 
217 
196 
173 
152 


276 
258 


265 
24fi 


0.81 
.88 
.86 

.81 

.78 


82 
76 
78 
70 
68 




11-12. . 




12-13 




252 235 


13-14 




221 
212 


208 
196 


14-15 


1st 


15-16 


2d 


159 


154 


211 


208 


.75 


66 


36.41 


16-17 


3d 


151 


148 


206 198 


.73 


62 




17-18 


4th ... . 


150 


140 


202 1 187 


.74 


65 


36.55 


18-19 


5th ... . 


143 


137 


192 


176 


.75 


63 


36.58 


19-20 


6th ... . 


134 


131 


194 


185 


.68 


60 


36.44 


20-21 


7th ... . 


135 


132 


190 


185 


.71 


59 


36.42 


21-22 


8th.... 


137 


135 


187 


177 


.73 


61 


36.55 


22-23 


9th.... 


134 


131 


178 


173 


.75 


59 


36.50 


23-24 


10th 


130 


127 


180 


179 


.72 


57 


36.64 


24-25 


11th.... 


128 


124 


176 


166 


.72 


57 


36.80 


25-26 


12th 


129 


126 


175 


173 


.73 


68 


36.85 


26-27 


13th 


126 


125 


171 


167 


.74 


56 


36.62 


27-28 


14th 


120 


116 


167 


160 


.72 


53 


36.30 


28-29 


15th 


117 114 


163 


162 


.71 


53 


36.43 


29-30 


16th 


117 


114 


165 


158 


.71 


53 


36.40 


Apr. 30-Mav 1 


17th.... 


115 


113 


160 


154 


.72 


52 


36.42 


Mav 1-2 


18th.... 


115 


112 


159 


154 


.72 


52 


36.30 


2-3 


19th.... 


113 


111 


158 


153 


.71 


52 


36.21 


3-4 


20th.... 


114 


112 


160 


159 


.71 


52 


36.51 


4-5 


21st 


112 


103 


154 


137 


.73 


54 


36.12 


5-6 


22d 


111 


109 


154 


153 


.72 


53 


36.10 


6-7 


23d 


112 


106 


156 


144 


.72 


66 


35.98 


7-8 


24th 


109 


106 


158 


152 


.69 


55 


35.88 


8-9 


25th .... 


111 


108 


153 


147 


.72 


55 


36.31 


9-10 


26th .... 


111 


106 


159 


151 


.70 


56 




10-11 


27th 


111 


107 


153 


145 


.72 


57 


36.03 


11-12 


28th 


115 


109 


162 


145 


.71 


59 


36.37 


12-13 


29th 


112 


104 


158 


152 


.72 


58 


36.23 


13-14 


30th.... 


110 


103 


151 


147 


.72 


58 


36.06 


14-15 


31st. . . . 


115 


109 


160 


148 


.72 


57 


36.14 


16-17 




124 

188 


117 
176 


154 
194 


143 
182 


.80 
.97 


64 
90 


36.79 
37.53 


17-18. . 









1 For the duration of the period during which the metabolism was measured see table 44. 
2 The duration of the periods in which these minimum values were observed varies in general 
from 3 hours to 1 hour. 



THE RESPIRATORY EXCHANGE. 329 

periodic values for the total gaseous exchange and the respiratory 
quotient for each night of the experiment. 

The average for the carbon-dioxide production varied from 228 c.c. 
per minute on April 11-12 (the second night of the preliminary food 
period) to 109 c.c. on May 7-8 (the twenty-fourth night of the fast). 
The carbon-dioxide excretion during the fasting period was much less 
than during the days when food was taken, ranging from 165 c.c. on 
the first night to 109 c.c. on the twenty-fourth night. This increase 
with the taking of food is also shown in the two nights following the 
fasting period, when the carbon-dioxide production increased from 115 
c.c. to 124 c.c. on the second night after the food was taken, and to 
188 c.c. on the third night. 

The minimum periods have a special interest, as they indicate the 
lowest plane of metabolism during the experiment. These values range 
from 218 c.c. on the first night with food to 103 c.c. on both the twenty- 
first and thirtieth nights of the fast. As with the average values, an 
increase after taking food is noted in the minimum periods for the two 
nights following the fast. 

While the carbon dioxide of itself is a well-known index of the metab- 
olism, particularly in fasting, when the character of the material 
burned remains relatively constant, nevertheless the values for oxygen 
likewise have significance. The average values for the oxygen consump- 
tion range from 276 c.c. on the first night with food to 151 c.c. on the 
thirtieth night of the fast. Considering only the fasting values, the 
oxygen consumed ranges from 212 c.c. on the first night of the fast to 
the minimum noted above. While no increase in the oxygen con- 
sumption is apparent in the first value obtained after the taking of 
food, a considerable increase is shown on May 17-18, the oxygen con- 
sumed being 194 c.c. 

The minimum periodic values for the oxygen consumption can not be 
considered to have the same degree of accuracy as the minimum 
periodic values for the carbon-dioxide production, but a comparison is 
of interest. These values range from 265 c.c. on the first night with 
food to 137 c.c. on the twenty-first night of the fast. During the fast- 
ing period the range is from 208 c.c. on the second night of the fast to 
137 c.c. on the twenty-first night. As with the average values, the 
minimum periodic values for the oxygen consumption do not show an 
increase after taking food until the third night. 

From an examination of all of the data given in table 45 for the car- 
bon-dioxide production and the oxygen consumption, it will be seen 
that there was a striking tendency for the total metabolism to decrease 
as the fast progressed. Both the average and the minimum periodic 
values show that the metabolism reached a low point about the 
twentieth day of the fast, and from that time to the end of the fast there 
was but little alteration. 



330 A STUDY OF PROLONGED FASTING. 

The oxygen consumption showed the same general course as the 
carbon-dioxide production, there being a steady decrease until about 
the twentieth day. While the values for the oxygen between the 
twenty-first and the thirty-first days do not show the regularity that 
was observed with the carbon dioxide during the same period, they still 
do not fluctuate widely from the low value, the average values remaining 
between 151 c.c. and 162 c.c. and the values for the minimum periods 
between 137 c.c. and 153 c.c. It will be seen, therefore, that the metab- 
olism as indicated by the carbon-dioxide production and the oxygen 
consumption decreased regularly until the twentieth day and from that 
time until the end of the fast remained essentially constant. 

This observation is strikingly significant, inasmuch as one would nat- 
urally expect that, as the organism wasted away as a result of fasting, 
the total metabolism would decrease and likewise the intensity of the 
metabolism. The fact that the decrease in the metabolism did not con- 
tinue beyond the twentieth day is the more surprising, since the loss 
in weight continued regularly throughout the fast. The absence of a 
continued decrease in the metabolism will subsequently be given special 
discussion. 

Respiratory Quotient. 

At present the best index we have of the character of the material 
burned in the body is the relationship between the volume of the carbon 
dioxide excreted and the oxygen consumed, i. e., the so-called respira- 
tory quotient. When carbohydrates are burned, the volume of carbon 
dioxide produced is equal to that of the oxygen consumed, the respira- 
tory quotient being 1.0. On the contrary, when fat is burned, there is 
a much less volume of carbon dioxide produced per liter of oxygen and 
the respiratory quotient is not far from 0.7. 

In the study of short fasts previously made at Wesleyan University, 
it was shown that the carbohydrates stored in the body (chiefly in the 
form of glycogen) were heavily drawn upon in the first few days of the 
fast and thereafter the body subsisted substantially upon fat and pro- 
tein, but chiefly fat. In this fasting experiment, therefore, a rapid 
fall in the respiratory quotient would be expected during the first days 
of the fast, and a possible constancy during the remaining days, show- 
ing a combustion of fat. 

By reference to the values for the respiratory quotient given in table 
45, it will be seen that on the 4 nights prior to the fast the values 
ranged from 0.81 to 0.88, averaging not far from 0.84. This quotient 
is approximately that which would be expected with individuals sub- 
sisting upon a mixed diet. On the first day of fasting, the respiratory 
quotient fell to 0.78 and remained for the next few days not far from 
0.74 to 0.75. On the sixth day a low value was found of 0.68, but for 
the remainder of the fast the respiratory quotient ranged above or 



THE RESPIRATORY EXCHANGE. 331 

below the average of 0.72. No average value lower than 0.68 was found 
in any of the experiments. These respiratory quotients would indicate 
that the combustion in the body after the first few days of fasting was 
principally of fat. As will be seen later, there was the formation of a 
small amount of /3-oxybutyric acid, which would have a tendency to 
lower slightly the respiratory quotient, but this would be in part com- 
pensated by the consumption of protein and a possible steady, though 
very small, oxidation of carbohydrate, both of which would tend to 
increase the respiratory quotient. Such an increase is indicated by the 
slightly higher average value of 0.72. 

The two nights after the fast, when food had been taken, show a 
marked increase in the respiratory quotient, the quotient for May 16- 
17 being 0.80. On the night of May 17-18, when the whole alimentary 
tract of the subject was filled with carbohydrate material, due to the 
excessive amount of fruit juices and honey which he had taken, the 
extraordinarily high value of 0.97 was obtained. 

The main conclusions to be drawn from the average respiratory 
quotients found in the experiments with the bed calorimeter as the fast 
progressed may be summed up as follows : 

First, no very low values were found, such as have been observed and 
reported by other investigators. Quotients below 0.68 were very 
rarely found for the individual periods, and the average value for the 
31 nights of the fast was 0.72. 

Second, from the course of the respiratory quotients, it is clear that 
carbohydrate was burned on the first few days of fasting, which is in 
full conformity with the results found in the experiments carried out at 
Wesleyan University. 

Finally, after the subject had fasted for 6 or 7 days, the respiratory 
quotients reached a point which indicated essentially a fat katabolism 
and continued at this point until the end of the fast, the formation of a 
small amount of /3-oxybutyric acid tending to lower the respiratory 
quotient and the combustion of a small amount of protein, with possibly 
a small amount of glycogen, tending to increase the quotient above that 
which would be obtained with the combustion of pure fat. 

Relationships of Pulse-Rate, Bodt-Tempebatube, and Metabolism. 

In considering both the curves of the respiratory exchange and the 
average values shown in table 45, it should be noted that two of the 
factors affecting the total metabolism were absent, i. e., muscular 
exercise and the digestion of food. Considering that the subject is 
living on a low metabolic plane, we might expect that the metabolism 
would be represented by a straight line, were it not for the influence of 
a third important factor — the internal muscular activity. The best 
index of the internal activity is the pulse-rate. Johansson 1 has also 

Johansson, Skand. Archiv f. Physiol., 1898, 8, p. 85. 



332 A STUDY OF PROLONGED FASTING. 

pointed out that there is an intimate relationship between the body- 
temperature and the metabolism. It is important, therefore, to con- 
sider the relationship between the metabolism as shown by the gaseous 
exchange and these two indices of the internal condition. 

It has seemed impracticable to complicate the curves in figures 41 to 
44 by superimposing others, but a comparison can be made by referring 
to the curves for the pulse-rate and the body-temperature 1 given in 
previous sections of this publication. Such a comparison shows that 
the curves for the carbon-dioxide excretion, the oxygen consumption, 
the pulse-rate, and the body-temperature have in general the same 
course for each experiment, with a distinct tendency to fall off during 
the evening until a minimum is reached about the middle of the night, 
and then to rise in the morning. This parallelism with the metabolism 
is shown more clearly in the curves for the body-temperature, as there 
are more variations in the curves for the pulse-rate, but the general 
rhythm of the latter is much like that exhibited by the curves for the 
metabolism. Furthermore, there does not appear to be a material 
difference in these two relationships at the beginning and end of the 
fast, so that it would seem that fasting per se does not affect them. 
The body acts as a unit, therefore, irrespective of the state of nutrition. 
The intimate relationship between the pulse-rate and the metabolism 
(which has been emphasized in this laboratory for a number of years) 
and the relationship between the body-temperature and the metabolism 
are thus not only demonstrated in a remarkable manner, but are also 
shown to be unaffected by a prolonged fasting period. 

The relationship between the pulse-rate and the total metabolism 
and the body-temperature and the metabolism as the fast progressed 
may be discussed more in detail in connection with the values given 
in table 45, using the average values rather than those for the minimum 
periods. In comparing these factors with the total metabolism on 
successive nights, it should be borne in mind that the relationships 
would not logically be expected to remain constant, for we have on the 
one hand the pulse-rate and the body-temperature governed by certain 
laws and on the other an organism producing heat, the heat-producing 
mechanism of which is constantly diminishing in size. 

The pulse-records for the nights preceding the fast are somewhat 
irregular, but the technique for making the observations was not then 
so perfectly developed as it was later in the experiment and the assis- 
tant had not the time to make such frequent records. It will be seen, 
however, that the high pulse-rates were obtained with the high values 
for the carbon dioxide and the oxygen during the 4 nights prior to the 
fast and throughout the first 2 weeks of the fasting period. In the 
latter portion of the fast there was a distinct tendency for the average 
pulse-rate to increase without a corresponding increase in the total 

1 See figures 4 to 8, pages 90 to 94, and figures 12 to 18, pages 104 to 110. 



THE RESPIRATORY EXCHANGE. 333 

carbon-dioxide output and oxygen intake. On May 17-18, however, 
the greatly increased pulse-rate was accompanied by a corresponding 
increase in both the carbon-dioxide output and the oxygen intake. 
In general, then, one may infer that even with an organism whose heat- 
producing mechanism is constantly decreasing in size, there is still 
an intimate relationship between the pulse-rate per minute and the 
total heat production. It should also be recognized that this relation- 
ship was somewhat disturbed during the latter portion of the fast, 
but not sufficiently disturbed as not to be again apparent on the third 
day with food. 

The body-temperature was not recorded on the nights preceding 
the fast, but observations were made nearly every night of the fast and 
for two nights following. The values given in table 45 for the fasting 
period have a distinct tendency to remain not far from an average 
value of 36.36° C, and range from 36.85° C. on the twelfth night of 
the fast to 35.88° C. on the twenty-fourth night. From the twenty-first 
night of the fast, the values for the most part lie distinctly below the 
average of the first 3 weeks of the fasting period; but little if any rela- 
tionship is shown between the average body-temperature and the total 
metabolism. On the other hand, on the last night of observation 
after the fast (May 17-18), the increased metabolism and increased 
pulse-rate were accompanied by the highest average temperature 
found on any night with this subject. 

It is evident, therefore, that while there is a tendency towards a 
parallelism of the body-temperature and the metabolism throughout 
any given night, there is no distinct tendency towards parallelism 
between the average temperatures of successive nights and the total 
metabolism as measured. Evidently the heat-regulating mechanism 
of the body is in large part independent of the total heat-production 
or of the condition of nutrition of the subject. 

STUDIES WITH THE UNIVERSAL RESPIRATION APPARATUS. 

The facility with which experiments could be carried out with the 
universal respiration apparatus made it specially adapted for measuring 
the metabolism of the fasting subject under various conditions, such as 
lying awake, sitting up either quietly or writing, or lying awake breath- 
ing an oxygen-rich atmosphere. It was also possible to obtain accu- 
rate determinations of the respiratory quotient with this apparatus. 
This was of special importance, since it was desired to establish as 
sharply as possible the respiratory quotient obtaining during prolonged 
fasting, particularly as the low quotients found by Luciani and by 
Zuntz and his co-workers have been the subject of much discussion. 
Consequently it was decided that, throughout the entire fast, respira- 
tion experiments would be made as frequently as practicable in which 
the respiratory exchange would be determined under various condi- 



334 



A STUDY OF PROLONGED FASTING. 



tions. A summary of the data obtained in these experiments is given 
in table 46. 

For purposes of comparison the oxygen absorbed and the carbon 
dioxide produced were calculated on the basis of cubic centimeters 
per minute. The respiratory quotient for each experiment and the 
average pulse-rate are also given in this table. The morning respira- 
tion experiments were made immediately following the calorimeter 



Table 46. — Gaseous exchange of subject L. at different times 
activity. (Respiration apparatus.) 


of the day and with varying 


Date. 


Day of 
fast. 


Lying (usually 8 h 30° a.m. to 
9 h 30™ a.m.). 


Lying (usually 7 p.m. to 
7 h 45 m p.m.). 


Carbon 
dioxide 

per 
minute. 


Oxygen 

per 
minute. 


Respi- 
ratory 
quo- 
tient. 


Aver- 
age 
pulse- 
rate. 


Carb 

dioxi 

per 

minu 


on --. 

de 0xygen 
per 

, minute. 
;e. 


R 

r: 
c 

ti 


espi- Aver- 
tory age 
[uo- pulse- 
ent. rate. 


1912. 




c.c. 


c.c. 






c.c. 


c.c. 






Apr. 11 




186 
196 
200 
182 
185 


231 
220 
225 
223 
237 


0.81 
.89 
.89 
.82 
.78 


72 
73 
72 
73 
74 












12 








13 




14 




15 


1st 


16 


2d 


180 


227 


.79 


73 














17 


3d 


169 


226 


.75 


70 














18 


4th 


159 


212 


.75 


68 














19 


5th ... . 


158 


205 


.77 


67 














20. . .. 


6th 


148 


200 


.74 


64 














21 


7th ... . 


153 


204 


.75 


64 














22 


8th 


151 


203 


.74 


65 














23 


9th.... 


143 


190 


.75 


63 














24 


10th.... 


143 


187 


.76 


63 














25 


11th.... 


140 


187 


.75 


61 














26 ... . 


12th 


140 


187 


.75 


61 


13< 


) 193 





72 62 


27 


13th.... 


140 


192 


.73 


59 


l i3( 


\ ! 195 


i 


70 »59 


28 


14th 


134 


181 


.74 


58 


134 


I 190 




71 59 


29 


15th 


132 


179 


.74 


57 


137 


189 




72 61 


30 


16th 


133 


182 


.73 


58 


134 


[ 190 




71 59 


May 1 . . . . 


17th 


130 


182 


.71 


57 


13C 


1 188 




69 61 


2 


18th.... 


123 


174 


.71 


56 


128 


1 189 




68 62 


3 


19th 


127 


177 


.72 


57 


12e 


182 




69 60 


4 


20th 


124 


173 


.72 


58 










5 


21st 


126 


174 


.73 


59 


12£ 


182 




69 57 


6.... 


22d 


124 


170 


.73 


59 


121 


176 




71 63 


7 


23d 


121 


165 


.73 


58 


12e 


175 




72 60 


8 


24th .... 


122 


167 


.73 


59 


12£ 


177 




71 63 


9 


25th 


125 


166 


.75 


60 


124 


176 




70 63 


10 


26th 


123 


168 


.73 


61 


128 


180 




71 66 


11 


27th.... 


129 


172 


.75 


62 


126 


181 




70 66 


12 


28th 


124 


166 


.75 


61 


123 


178 




69 66 


13 


29th.... 


124 


171 


.73 


63 


123 


178 




69 67 


14 


30th .... 


119 


166 


.72 


59 


127 


183 




69 71 


15.... 


31st 


120 


166 


.72 


60 










17 




133 
172 


170 
183 


.78 
.94 


72 
84 










18 

















1 During a period from 3 h 16 m p.m. to 3 h 51 m p. m. on this day, with the subject in the lying 
position, the observations were: Carbon dioxide, 140 c.c; oxygen, 189 c.c; respiratory quotient, 
0.74; pulse-rate, 61 per minute. 



THE RESPIRATORY EXCHANGE. 



335 



Table 46. — Gaseous exchange of subject L. at different times of the day and with varying 
activity. {Respiration apparatus.) — Continued. 



Date. 


Day of 

fast. 


Sitting. 1 


Period. 


Carbon 
dioxide 

per 
minute. 


Oxygen 
per 

minute. 


Respira- 
tory 
quotient. 


Average 
pulse- 
rate. 


1912. 

Apr. 16. . . 

19... 

23... 

24... 

26... 

27... 

29... 

May 1 . . . 

4... 

7... 

14... 


2d 

5th.... 

9th.... 

10th 

12th 

13th.... 

15th.... 

17th.... 

20th.... 

23d 

30th 


4 h 00 m p.m. to 4 b 35 m p.m. 
4 10 p.m. 4 43 p.m.* 
3 52 p.m. 4 28 p.m. 
3 58 p.m. 4 57 p.m. 
3 13 p. n. 4 11 p.m. 
12 14 p.m. 12 48 p.m. 
3 23 p.m. 3 56 p.m.* 
9 31 a.m. 10 04 a.m.* 
9 35 a.m. 10 10 a.m.* 
3 43 p.m. 4 14 p.m.* 
6 32 p.m. 7 02 p.m.* 


c.c. 

179 

198 

129 

144 

124 

141 

164 

153 

141 

159 

156 


c.c. 

244 

269 

187 

194 

183 

200 

233 

215 

208 

222 

221 


0.73 
.74 
.69 
.74 
.68 
.71 
.70 
.71 
.68 
.72 
.71 


82 
80 
62 
69 
60 
68 
68 
69 
65 
69 
75 



1 Periods indicated by an asterisk were obtained with the subject sitting, writing. 

experiment, with the subject still lying upon the couch in essentially 
the same position as when he left the calorimeter chamber. The 
evening respiration experiments were made in the latter days of the 
fast just before the subject entered the calorimeter chamber. The 
experiments when the subject was sitting were of two kinds. In certain 
experiments he sat quietly in his chair, but in others he was writing, 
exactly as is shown in plate 1, figure b. It is thus seen that observa- 
tions were made with this fasting subject on every day of the fast and 
on certain days experiments were made in several body positions. 
These data also give an indication of the diurnal variations in the katab- 
olism, since observations were made under identical body conditions, so 
far as muscular activity and absence of food are concerned, both in 
the morning after the subject left the calorimeter and in the evening 
on the same day just prior to entering the chamber for the night. 



VARIATIONS IN THE METABOLISM AS THE FAST PROGRESSED. 

The carbon-dioxide production of the subject while lying on the 
couch in the morning respiration experiments ranged from 200 c.c. 
on the morning of April 13 (one of the days preceding the fast) to 119 c.c 
on May 14 (the thirtieth day of fasting). During the fasting period 
the values ranged from 185 c.c. to 119 c.c. When the subject again 
took food the rise in the carbon-dioxide production was very notice- 
able, particularly on the last day of observation. 

In general, the data obtained for the oxygen consumption nearly 
paralleled those for the carbon-dioxide production. The maximum 



336 A STUDY OF PROLONGED FASTING. 

value of 237 c.c. was obtained on the first day of the fast, while the 
minimum value of 165 c.c. was found on the twenty-third day of the 
fast. The striking constancy shown in the values for the oxygen con- 
sumption on the first four mornings prior to the fast and on the first 
day of the fasting period is worthy of special notice, as it gives evidence 
in the first place of the remarkable constancy in the katabolism of this 
man and likewise of the regularity of his muscular repose. Both the 
carbon-dioxide production and the oxygen consumption fell off as the 
fast progressed, and although the minimum value for the oxj'gen con- 
sumption was reached on the twenty-third day, yet the remaining days 
of the fasting period indicate a katabolism not far from the minimum 
value of 165 c.c. It is of particular interest to note that on the 17th 
of May, i. e., the second day of taking food, the metabolism had not 
materially increased, as shown by the oxygen consumption, but on the 
last morning (May 18) there was a marked increase from 170 c.c. to 
183 c.c. 

The trend of the respiratory quotient is likewise significant. On the 
first 4 days with food, the respiratory quotient varied from 0.81 to 0.89, 
this being not far from the average respiratory quotient found with 
normal individuals subsisting on a mixed diet. At the beginning of the 
fast, the respiratory quotient was a little lower on the first few days 
and then steadily decreased until a minimum value of 0.71 was found 
on the seventeenth and eighteenth days. During the remainder of 
the fast, the value for the respiratory quotient remained at about 0.73. 
On the second day with food it rose to 0.78, and on the third day with 
food it reached the extraordinarily high value of 0.94, indicating that 
the subject was surcharged with carbohydrate material. The absence 
of very low quotients during the fast was noticeable. It should be 
borne in mind that the values for the oxygen consumption represent 
more nearly the true index of the total metabolism than do the values 
for the carbon-dioxide production, particularly in the 4 days with food 
preceding the fast and the first few days of fasting. After the third or 
fourth day of fasting, however, the values for the carbon-dioxide pro- 
duction and the oxygen consumption were essentially parallel, so that 
either may be looked upon as a true measure of the total metabolism. 

RELATIONSHIP BETWEEN THE PULSE-RATE AND THE METABOLISM. 

The pulse-rate remained remarkably constant for the first 4 days 
with food before the fast and likewise on the first few days of fasting, 
ranging from 72 to 74, which is in general conformity with the measure- 
ments of the oxygen consumption. It then fell with a considerable 
degree of regularity until a minimum value of 56 was reached on the 
eighteenth day of the fast. Subsequently the values show a slight, 
though definite, tendency to rise gradually to the end of the fasting 
period. The increase on the second day after food was taken was 



THE RESPIRATORY EXCHANGE. 337 

considerable, with a still further increase on the last day on which the 
observations were made. 

A careful examination of the fluctuations in the values for the oxygen 
consumption and the pulse-rate shows a remarkable regularity in the 
relationship between them, although the absolute minimum values were 
not found on the days that the minimum pulse-rate was found. On 
the second day with food after the fast, the pulse-rate rose to 72 and 
the oxygen consumption likewise rose, reaching 170 c.c. The values 
taken as a whole, however, show that in the earlier days of this long 
fast the relationship between the oxygen consumption and the pulse- 
rate was reasonably close, but in the latter part of the fasting period 
there was a slight divergence, as a somewhat increased pulse-rate was 
occasionally accompanied by an actual decrease in the oxygen con- 
sumption. It should be considered, however, that the organism was 
changing from day to day, and while the total tissue available for 
metabolism was slowly decreasing the pulse-rate may still have a 
definite relationship to the total active tissue remaining. Thus a 
decrease in the amount of tissue may in part be compensated for by an 
increase in the pulse-rate, although this latter factor may still have too 
small an effect to prevent a lowering of the total metabolism. Further 
discussion along this line must be deferred until the metabolism per 
unit of body-weight and per unit of body-surface are considered. 

DIURNAL VARIATIONS IN METABOLISM. 

The determination of the respiratory exchange at various times 
during the day gives an excellent opportunity for studying the diurnal 
variations in the metabolism of the same individual during fasting. 
The data given in table 46 show that the metabolism during the evening 
experiments was invariably higher than in the morning experiments, 
regardless of whether the carbon-dioxide production or the oxygen 
consumption is used as an index. 

The pulse-rate was also a few beats higher in the evening hours, thus 
indicating a close relationship between the pulse-rate and the metab- 
olism. The slight tendency for the pulse-rate to rise in the morning 
experiments beginning with the eighteenth day of the fast and con- 
tinuing to the end was even more marked in the records of the pulse- 
rate for the series of evening experiments, in which the minimum value 
of 57 was found on the twenty-first day of the fast and the maximum 
of 71 on the thirtieth day. A general relationship between the oxygen 
consumption and the pulse-rate is shown throughout all of the series 
of experiments, although as the fast progressed this relationship was 
not so pronounced as at the beginning. It should be considered here 
again, however, that the active mass of protoplasmic tissue was gradu- 
ally decreasing, so that the relationship can not be expected to hold 
constant. 



338 A STUDY OF PROLONGED FASTING. 

The respiratory quotients obtained in the evening were not unlike 
those obtained in the morning experiments, with a slight tendency for 
the early evening quotients to be somewhat lower than those obtained 
in the morning experiments. This lowering of the quotient in the 
evening experiments might be taken as an indication that there may 
have been a formation of carbohydrate from fat by a storage of oxygen, 
or a greater formation of /S-oxybutyric acid, and that the next morning 
either the formation of the /3-oxybutyric acid was less or that the slight 
supply of glycogen formed during the early evening was being burned. 
Unfortunately, although these respiratory quotients were determined 
with the best technique that we know of at present, we do not feel 
justified in laying much stress upon a change of one or two units in the 
quotients. 

EXTERNAL INFLUENCES UPON METABOLISM. 

While the values for the carbon-dioxide output, the oxygen intake, 
the respiratory quotient, and the pulse-rate are given in table 46 for the 
experiments in which the subject was lying upon a couch and sitting 
up in a chair, either writing or quietly at rest, a better understanding 
of the influence of a change in conditions may perhaps be secured by 
studying each change by itself. To this end several small tables have 
been prepared which show the influence of the change in condition 
upon the total metabolism and also upon the mechanics of respiration. 

Effect of Changes in Body Position. 

On the second, tenth, twelfth, and thirteenth days of the fast, the 
metabolism was studied while the subject was sitting in a chair. It was 
thus possible to compare the metabolism and the mechanics of respi- 
ration for the two positions. This comparison is made in table 47, in 
which is given the increase or decrease in the values due to the change 
to the sitting position. The figures show that in general there was 
practically no increase in the carbon-dioxide production — and, indeed, 
in two instances a considerable decrease. The oxygen consumption 
was increased in 3 out of the 5 experiments, with a slight decrease in 
the other 2, and there was a perceptible though probably not significant 
change in the respiratory quotient, which may have been caused by 
the absence of change in the carbon-dioxide production. There was 
an average increase in the pulse-rate and respiration-rate and an 
increase in the lung ventilation, but varying results in the volume per 
respiration. 

Since the lying experiments were made in the early morning and the 
sitting experiments late in the afternoon, they are not, strictly speaking, 
comparable. On the other hand, one would expect that there would be 
a higher metabolism normally in the late afternoon than in the morning 
after the subject came out of the calorimeter, and it is accordingly very 



THE RESPIRATORY EXCHANGE. 



339 



difficult to explain the results obtained on the ninth and twelfth days of 
the fast, when there was an actual decrease in the oxygen consumption, 
with a slight falling off of the pulse-rate. 

The general course of the metabolism noted on the second, tenth, 
and thirteenth days is essentially that which is found with normal 
individuals — namely, a small increase due to the position of sitting. 
This increase was also accompanied by an increased pulse-rate. The 
parallelism shown here between the increase of the oxygen consumption 
and the pulse-rate is worthy of special attention. 

The figures also show that the change to the position of sitting 
invariably results in an increased ventilation of the lungs, although 
the respiration-rate changes so that the actual volume per inspiration 
may be above or below that when the subject was lying. No positive 
deductions can be drawn as to the influence of the change in position 
upon the volume per inspiration. If an average of these five experi- 
ments were permissible, it would be seen that there was an increase in 
metabolism of not far from 5 c.c. per minute, or about 2 to 2.5 per cent 

Table 47. — Comparison of the gaseous exchange and lung ventilation of subject L., lying 
on couch and sitting in chair. (Respiration apparatus.) 



Date. 


Day 
of 

fast. 


Position. 


No. 
of 

peri- 
ods. 


Time. 


Car- 
bon 

diox- 
ide 
per 

min- 
ute. 


Oxy- 
gen 
per 
min- 
ute. 


Respi- 
ratory 
quo- 
tient. 


Respi- 
ration 
rate. 


Lung 
venti- 
lation 
per 
min- 
ute. 1 


Vol- 
ume 
per 
inspi- 
ration. 2 


Pulse- 
rate. 


1912. 
L .pr. 16 

. .pr. 23 

. .pr. 24 

. v.pr. 26 

'..pr.27 


2d 
9th 
10th 
12th 
13th 


Lying 

Sitting .... 


3 
2 


8 h 34 m a.m. to 9 h 37 m a.m. 
4 h 00™ p.m. to 4 h 35 m p.m. 


C.C. 

180 
179 


c.c. 
227 
244 


0.79 
.73 


10.9 
10.3 


liters. 
5.18 

5.58 


C.C. 

576 
660 


73 

82 


-1 


17 




-.6 


.40 


84 


9 


Lying 

Sitting .... 


3 

2 


8 h 25 m a.m. to 9 h 18 m a.m. 
3 h 52 m p.m. to 4 h 28 m p.m. 


143 
129 


190 

187 


0.75 
.69 


12.1 
16.7 


4.65 
5.48 


476 
402 


63 
62 


-14 


-3 




4.6 


.83 


-74 


-1 


Lying 

Sitting .... 


3 
3 


8 h 19 m a.m. to 9 h 37 m a.m. 
3 h 58 m p.m. to 4 h 57 m p.m. 


143 
144 


187 
194 


0.76 
.74 


10.9 
14.6 


4.55 
5.83 


504 

484 


63 
69 


1 


7 




3.7 


1.28 


-20 


6 


Lying 

Sitting .... 


3 
3 


8 h 21 m a.m. to 9 h 26 m a.m . 
3 h 13 m p.m. to 4 h ll m p.m. 


140 
124 


187 
183 


0.75 

.68 


12.8 
15.8 


4.64 
5.37 


429 
404 


61 
60 


-16 


-4 




3.0 


.73 


-25 


-1 


Lying 

Sitting . . . . 


3 

2 


8 h 37 m a.m. to 9 h 33 m a.m . 
12 h 14 m p.m.to 12 h 48 m p.m. 


140 
141 


192 
200 


0.73 
.71 


12.8 
12.8 


4.63 
5.55 


437 
525 


59 

68 


1 


8 




0.0 


.92 


88 


9 









^he lung ventilation observed is here reduced to 0° C. and 760 mm. pressure. 
Calculated to the pressure existing in the lungs and to 37° C. 






340 



A STUDY OF PROLONGED FASTING. 



of the oxygen consumption. In the light of these varying results, it is 
to be regretted that further observations were not made with the sub- 
ject sitting quietly. However, a number of observations made when 
the man was sitting up and writing actively may also be compared. 

Influence op the Work op Writing. 

On the fifth, fifteenth, seventeenth, twentieth, twenty-third, and 
thirtieth days of the fast, the metabolism was studied while the subject 
was sitting up writing, an employment that occupied much of his spare 
time during the entire fast. These six experiments are compared in 
table 48 with data obtained on the same day when the subject was 
lying upon the couch in the morning experiment. Two of these experi- 
ments — those on the seventeenth and twentieth days — immediately 

Table 48. — Comparison of the gaseous exchange and lung ventilation of subject L., lying 
on couch and sitting writing. (Respiration apparatus.) 



Date. 


Day 
of 

fast. 


Position. 


No. 
of 
peri- 
ods. 


Time. 


Car- 
bon 
diox- 
ide 
per 
min- 
ute. 


Oxy- 
gen 
per 
min- 
ute. 


Respi- 
ratory 
quo- 
tient. 


Respi- 
ration- 
rate. 


Lung 
venti- 
lation 
per 
min- 
ute. 1 


Volumt 

per 
inspi- 
ration.' 


Pulse 
rate. 


1912. 
Apr. 19 

Apr. 29 

May 1 

May 4 

May 7 

May 14 


5th 
15th 
17th 
20th 
23d 
30th 


Lying 

Writing. . . 

Increase 

Lying 

Writing . . . 

Increase 

Lying 

Writing. . . 

Increase. 


4 
2 


8 h 21 m a.m.to 9 h 32 m a.m. 
4 h 10™ p.m. to 4 h 43 m p.m. 


ex. 
158 
198 


c.c. 
205 
269 


0.77 
.74 


11.8 
17.9 


liter 8. 
4.88 
7.54 


c.c. 
507 
517 


67 
80 


40 


64 




6.1 


2.66 


10 


13 


3 

2 


8 h 19 m a.m.to 9 h 19" 1 a.m. 
3 h 23 m p.m.to 3 h 56 m p.m. 


132 
164 


179 
233 


0.74 
.70 


12.3 
18.7 


4.55 

7.88 


446 
510 


57 
68 


32 


54 




6.4 


3.33 


64 


11 


3 

2 


8 h 22 m a.m.to 9 h 1&» a.m. 
9 h 31 m a.m. to 10 h 04 m a.m. 


130 
153 


182 
215 


0.71 
.71 


12.3 
14.6 


4.81 
6.57 


471 
542 


57 
69 


23 


33 




2.3 


1.76 


71 


12 


Lying 

Writing . . . 

Increase. 


3 
2 


8 h 22 m a.m. to 9 h 15™ a.m. 
9 h 35 m a.m. to 10 h 10" a.m. 


124 
141 


173 

208 


0.72 
.68 


14.3 
15.3 


4.90 
6.22 


413 
490 


58 
65 


17 


35 




1.0 


1.32 


77 


7 


Lying 

Writing. . . 

Increase. 


3 
2 


8 h 15 m a.m.to 9 h 17 m a.m. 
3 h 43 m p.m. to 4 h 14 m p.m. 


121 
159 


165 
222 


0.73 
.72 


14.0 
16.1 


4.76 
7.62 


410 
573 


58 
69 


38 


57 




2.1 


2.86 


163 


11 


Lying 

Writing . . . 


3 
2 


8 h 03 m a.m. to 8 h 56 m a.m. 
6 h 32 m p.m.to 7 h 02 m p.m. 


119 
156 


166 
221 


0.72 
.71 


14.8 
17.8 


4.80 
8.05 


391 
546 


59 
75 


37 


55 




3.0 


3.25 


155 


10 












x The lung v 
Calculated 


sntilat 
to the 


ion observed is here reduced 
pressure existing in the lung 


to0°< 
i and 1 


3. and 
,o37° 


760 m] 
C. 


n. pres 


sure. 







THE RESPIRATORY EXCHANGE. 341 

followed the morning experiments, and it is again to be regretted that 
this routine was not carried out in all cases. 

There was a noticeable increase in the metabolism in all of the writing 
experiments, which is shown by both the carbon-dioxide excretion and 
the oxygen consumption, but the respiratory quotient tended to become 
a few points lower during the writing period. The pulse- and respira- 
tion-rates were both invariably increased, the increase in the pulse-rate 
ranging from 7 to 16 beats per minute. The ventilation of the lungs 
per minute likewise increased very considerably during the writing 
experiment, this increase at times amounting to 3£ liters. The volume 
per inspiration did not increase materially, except on the twenty-third 
and thirtieth days of the fast. 

Under ordinary conditions one would normally expect an increased 
metabolism in the afternoon over the morning, but the comparison 
previously made between the metabolism for the lying and sitting posi- 
tions showed no increase for the sitting position in some instances and 
in others there was an actual decrease. Consequently a sharp compar- 
ison is difficult to make for the writing experiments. 

We may assume from these experiments, however, that when the 
subject was writing there was invariably an increased metabolism, but 
that on the 2 days when the writing experiment immediately followed 
the lying experiment the increase was only 50 to 75 per cent of that 
obtained on the days when the writing experiment was in the afternoon. 
On the seventeenth and twentieth days of fasting, when the writing 
experiments were in the morning, the increase in the metabolism due to 
writing was represented by an increased consumption of about 35 c.c. 
of oxygen, or not far from 20 per cent. The absolute maximum increase 
of 64 c.c. above the lying position on the fifth day of the fast amounted 
to approximately 30 per cent, but while the absolute increase on the 
twenty-third and thirtieth days was a few cubic centimeters less, the 
percentage increase was greatest, i. e., 34 and 33 per cent respectively. 

The work of writing, therefore, produced a distinct increase in the 
metabolism, which is shown not only by the increase in the carbon- 
dioxide production and the oxygen consumption, but also by an 
increase in the respiration-rate, the ventilation of the lungs, and the 
volume per inspiration. Finally, there was a regular and distinct increase 
in the pulse-rate. These values are used subsequently in computing 
the probable metabolism of this subject during several hours in the 
day when he sat in the balcony and wrote. 

Influence of Breathing an Oxygen-rich Atmosphere. 

On the twenty-eighth, twenty-ninth, and thirtieth days of the fast 
a series of experiments was made in which the subject breathed an 
oxygen-rich atmosphere varying from 95 to 75 per cent of oxygen. 
At the beginning of each experiment the percentage of oxygen in the 



342 



A STUDY OP PROLONGED FASTING. 



atmosphere was probably not far from 95 per cent. This fell off quite 
rapidly through the experimental period, so that at the end of the 
experiment the proportion of oxygen in the atmosphere was between 
75 and 80 per cent. The results of these experiments are not included 
in table 46, but are compared in table 49 with the morning experiments 
in which the subject breathed a normal atmosphere. In each case 
the oxygen experiment immediately followed the regular morning 
experiment. 

Table 49. — Comparison of the gaseous exchange and lung ventilation of subject L., breath- 
ing different air mixtures. (Respiration apparatus, subject lying, in the morning.) 



Date. 


Day 
of 

fast. 


Air mixture. 


No. 
of 
peri- 
ods. 


Car- 
bon 
diox- 
ide 
per 
min- 
ute. 


Oxy- 
gen 
per 
min- 
ute. 


Respi- 
ratory 
quo- 
tient. 


Alveolar 

carbon 

dioxide. 


Respi- 
ration- 
rate. 


Lung 
venti- 
lation 
per 
min- 
ute. 1 


Volume 

per 
inspi- 
ration. 2 


Pulse- 
rate. 


1912. 
May 12 

May 13 

May 14 


28th 
29th 
30th 


Oxygen-rich. . . 
Increase. . . . 

Oxygen-rich. . . 
Increase. . . . 

Oxygen-rich. . . 
Increase. . . . 


3 
2 


c.c. 
124 
137 


c.c. 
166 
179 


0.75 

.77 


p. cl. 
3.66 
3.43 


14.8 
13.6 


liters. 
5.04 
5.50 


c.c. 
410 

487 


61 
62 




13 


13 




-.23 


-1.2 


.46 


77 


1 


3 
2 


124 
130 


171 
176 


0.73 

.74 


3.67 
3.34 


14.1 
14.0 


4.96 
5.44 


426 
471 


63 
61 




6 


5 





-.33 


-.1 


.48 


45 


-2 


3 
2 


119 
130 


166 
169 


0.72 

.77 


3.80 
3.44 


14.8 
14.2 


4.80 
5.34 


391 
454 


59 

58 




11 


3 




-.36 


-.6 


.54 


63 


-1 



*The lung ventilation observed is here reduced to 0° C. and 760 mm. pressure. 
Calculated to the pressure existing in the lungs and to 37° C. 

This series of experiments maybe compared with an extensive research 
made in this laboratory by Mr. H. L. Higgins on the influence upon 
the metabolism of normal individuals of breathing oxygen-rich atmos- 
pheres. In the previous experimenting it was found that there was no 
evidence of the increased percentage of oxygen affecting the respiratory 
exchange, but there was a slight tendency for the pulse-rate to decrease 
with an increase in the percentage of oxygen. In the experiments with 
L., observations were made of the effect on the carbon-dioxide excre- 
tion, oxygen consumption, respiratory quotient, the respiration-rate, 
lung ventilation, volume per inspiration, alveolar carbon dioxide, and 
the pulse-rate. 

The data given in table 49 show a slight increase in the metabolism 
indicated by both the carbon-dioxide production and the oxygen con- 
sumption. The respiration-rate had a tendency to fall off somewhat; 



THE RESPIRATORY EXCHANGE. 343 

the lung ventilation was considerably increased — the increase ranging 
from 0.46 liter to 0.54 liter — and the volume per inspiration was like- 
wise increased from 45 c.c. to 77 c.c. per inspiration. The pulse-rate 
was but little affected. 

It is clear, therefore, that, even with an emaciated subject after a 
prolonged fast, the inhalation of an oxygen-rich atmosphere did not 
materially affect the carbon-dioxide production and oxygen con- 
sumption, the slight increase being not far from 7 c.c. or approximately 
5 per cent. The increase in the lung ventilation was, however, con- 
siderable, amounting to about 10 per cent. This has special signifi- 
cance with reference to the clinical use of oxygen, since it has been 
contended that one of the advantages in the clinical administration 
of oxygen was the fact that the patient was able to secure a sufficient 
amount of oxygen for the aeration of the blood with less effort than 
would be required to pump the lungs full of ordinary room air. The 
larger ventilation of the lungs in the oxygen experiments would imply 
that the increased amount of oxygen from inhaling an oxygen-rich 
atmosphere was obtained with a greater mechanical effort, which would 
thus offset the supposed advantage of this method of securing a suitable 
percentage of oxygen for the oxygenation of the blood. 

Influence of Sleep. 

The measurements of the metabolism in the bed-calorimeter experi- 
ments were made for the most part when the subject was resting very 
quietly. During a portion of the time he was asleep; it is equally 
certain that during certain per